I understand that a lot of people really enjoy managing teams of agents and are far more productive now. I am happy for them, I really am. Apparently they never actually enjoyed coding, they will show up in droves in these comments to tell you that what they really enjoyed was the end result.
I am more productive too, but I'm going to be honest - I wasn't really looking for more productivity. I am at a point in my life where boosting my productivity is absolutely not one of my personal objectives.
Using Claude Code all day is like being a STEM kid that was gifted a Lego-building-robot. What a miracle! Now you don't have to deal with all the hassle of assembling Legos - you just tell it you want a castle and boom! What a time to be alive!
yep. ive accepted it, called it a good run, put away all my fun personal projects, and started pursuing a nursing degree i can treat as a stable day job
with code all i wanted to do was code. with a nursing at least ill want to leave it at the hospital (yes nurse ptsd is real but nurse pay is great in pnw)
i think it might be good to have a job i treat as a job and not a passion i carry with me everywhere
True, but a transformer system regresses to the mean. I don’t mind that with my bank’s app (in fact thats probably a good thing) but there are other kinds of applications where creativity will be required.
I really doubt we’ll have systems that come up with the Apple folding phone’s “blurry image of the outside screen as you open that snaps into focus as you finish opening it” feature, at least for a very long time. The very training system of these foundation models works against it.
At the time people didn’t understand that — they thought photography would kill painting. Then for a long time photography wasn’t considered an art form. Your comment reflects an understanding/ viewpoint that is much less than a century old.
We are currently at the point equivalent to when photography was just appearing (so to speak)
I find it entirely irrelevant that contemporary comments on emerging technology of photography may have in part worried about the future of representative arts. 19th century featured a lot of blowhard know nothing know it alls.
As for today, we are discussing distinct means of producing the same thing: software. Painting remains because photography was at most displacing a subset of the field. Generative AI writing software may be displacing an entire profession.
I feel incredibly productive. I'm building in an afternoon what would've taken weeks. I spend all my time in either MR's reviewing code, or a terminal prompting.
I've moved from being a coder, to being a product-engineer-architect. I love coding -- LOVE it. But the role has morphed. And, I actually really like the new role. I love being product centric, and caring about architecture -- it suits me wel..
The code Claude produces is often clean, but terribly archictected. So I'm spending a lot of time reviewing in detail.
But the anxienty - OMG the anxiety.
I work on multple features (generally 4-5 work streams) concurrently, so am constantly context switching. I worry about how much time I'm spending on each task. Am I doing enough code review? Too much? Too little?
What about click-through testing? Our test coverage has never been higher - should I be able to just merge these features? Wasn't that always the promise of TDD - merging with confidence? But, it feels like a risk.
And, I'm moving really really fast - but I assume my competitors are too. So that drives me to move faster. Automate more. Use claude in more places. Which brings more anxiety - am I cutting too many corners? Is this all gonna crumble on me?
Also, is everything I'm working on going to become irrelevant? Will customers buy this product, or vibe code it? Do we even have a moat? Anxiety. More anxiety.
My investors are reading stuff constantly too. They're pinging me with market updates, emerging players, shifting landscapes. More anxiety. Also - productivity tools - look - internet influencer X is now producing a gajillion lines of code a second -- are we doing that? Are we token maxing? Are we feature maxing?
Are we building the right number of internal tools? Are we automating enough? Should I evaluate the market for this new tool, or just smash something out with claude? More features. Burn more tokens. Ship ship ship.
Having to explain to investors why we're not going as fast as those guys, and building enterprise products requires care and dedication -- that takes time. Time I should be shipping, coding, featuring, tokenizing. More anxiety.
I'm exhausted.
Anyway, gotta go - herdr has pinged me four times during writing this reply. Gotta ship.
They're representative of scenarios which your application can enter.
So if the tests were mocked and a bug went live he'd still notice, after all it means your software behaved incorrectly and now you gotta clean up the pieces.
This is becoming cyclical, did you forget what this comment chain went about?
The discussion went from tests being written by ai, to nobody actually reviewing the tests, to the claim that given those circumstances, you wouldn't notice that the tests may not tests anything anymore because of mocks.
My statement was merely pointing out that your previous comment was groundless because they still would've found out about those bugs when they went live.
>My statement was merely pointing out that your previous comment was groundless because they still would've found out about those bugs when they went live.
That's a big issue in my team. We cannot really review tests manually, we already have too much just reviewing the code. And even then, sometimes the 'is this change necessary' question is forgotten or ignored (happened to me 20 minutes ago) because I an tired of fighting my junior+AI. I try to make AI refactor the tests every weeks ande that's it.
> Are people reviewing all the tests that have been created automatically?
Absolutely not! Some people ask a code assistant to write only integration/end-to-end tests and commit them. Unittests is only to help LLM generate correct code.
I spend about 90% of at-computer time in a herdr-like tool.
It feels like playing factorio: whatever is manual needs to be automated away.
LLM writes the test. The test is crap. Update the skill. Ask an LLM to review your feedback on the last 5 PRs and write a skill for a review bot to look for similar problems and post. Now you only see PRs after they've gone through a few rounds of review.
The LLM has stopped picking up new issues. Ask an LLM to diagnose why. Ask an LLM to write a cron to regularly check if the LLM PR pipeline has halted and ping you. Get an LLM to try to diagnose it first.
I'm using a lot of LLMs! Ask an LLM to diagnose why. Write a tool to audit your usage, and produce a weekly report. Have it raise issue to reduce spending.
And like a base building game there are many competing facets, coming all at once.
Haha, I have been building my own software factory exactly like that. I consider it the ultimate factory game- but I sure as fuck would not want that to be my dayjob... Though I'm aware it will likely be that in a few years when the industries in my nation normalize around that.
No offense, but this feels like the most un-Factorio thing imaginable.
In factory building games, the systems in play can be understood and are fully deterministic. You're given a set of tools and a problem, and then craft together a solution. Your reward isn't just seeing the item you want on a conveyor belt, but looking at the factory you built ticking like clockwork.
What you're describing is beating out endless nondeterministic fires. It's like you were forced to play Factorio exclusively through an algorithm that designs incomprehensible factories that don't work sometimes, but designs them much faster than you ever could. So you click the magic button and see a mass of buildings appear in front of you. They may work perfectly fine, but if they don't, you need to use the same statistical factory estimator to build an in-game factory fixer mechanism and factory tester mechanism. And then a few updates later the factory fixer breaks so you need to deploy a factory fixer fixer, and a belt clean-up machine, and so on ad infinitum. To me this doesn't seem satisfying in the slightest.
I don't really see how it is as different as you say. Even if a system is stochastic, it can still be controlled, it's just a different type of engineering. Much of traditional engineering works exactly like this, with tolerances and probability rather than absolute determinism.
I think we (software engineers) are suffering from partial blindness to this aspect of engineering given that software has always been so extremely deterministic. We've trained ourselves that solving a problem means composing small fully deterministic parts into deterministic systems, and the prospect of having to work more with statistical confidence and managing failure modes feels scary and wrong.
No matter what your opinion on determinism is, the point I made that what OP is doing is very unlike a predictable and deterministic factory game still stands.
But beside that, I don't think that LLM non-determinism is really the same thing as what you see in engineering. Engineering has things like tolerances and redundancy, and while they accommodate random events, they're known, estimated and expected. They also deal with external factors, not the potential of shoddy engineering in the same system they're working on. The best way I can compare it is building a bridge that can withstand twice the normal expected full load and with enough structural redundancy to not collapse even after a prolonged lack of maintenance. These things account for many types of natural degradation and edge cases that are expected. And then there's building a bridge that you know will collapse in the next 10 years at a completely random moment, and instead of engineering a more robust solution, tacking on more gadgets to make the collapse less bad - automatic gates that close the entrances to the bridge, lake drainage system that will suck out the debris, a drone system that drops a brand new randomized bridge on the same spot, and so on. It doesn't feel very much like engineering.
If that doesn't sound convincing to you, consider one other thing. If you gave an engineer a non-deterministic solution and a deterministic solution to the exact same problem, which do you think they would prefer? Would they willingly remove determinism from a system? Maybe it's not the chaos of the real world that scares software developers so much - lots of them have to work on systems that directly interact with the real world. What if it's the fact that there is now a fight to take away their nice, understandable, satisfying deterministic solutions and inject probabilistic chaos into things that never had to have it in the first place - into absolutely everything?
> No matter what your opinion on determinism is, the point I made that what OP is doing is very unlike a predictable and deterministic factory game still stands.
I didn't mention determinism or predictability.
I said "It feels like playing factorio: whatever is manual needs to be automated away.".
I'm sure you can find many differences between Factorio and what I'm doing.
> If you gave an engineer a non-deterministic solution and a deterministic solution to the exact same problem, which do you think they would prefer? Would they willingly remove determinism from a system?
We choose non-determinism in our systems a lot, because we get better outcomes in other aspects.
* Non-deterministic network latency: Packet switching rather than circuit switching.
* Multi-tasking operating contending for the same resources.
* Multi process programs whose interleaving/execution is non-deterministic.
* Profile guided optimisation, depends on the profile.
* The entire field of randomised algorithms.
I prefer deterministic processes, but if the benefits of non-deterministic processes are significant enough, I pick non-deterministic.
Looks like written by an LLM, but it gives a realistic picture. The investors breathing down your neck is something that you don't want in this age, though.
- spending more time upfront working with agent on intent of the change, then design, so agent does not try to work around issues in a way that I don't intend to
- investing in ways in which agent can verify it's own change
- giving access to as much prod telemetry, etc so it can also verify the change post-merge
- obvious stuff, like automated review loops etc
But in the end I still read the "load bearing" parts of the code. Less closely, and I've long ago stopped looking at the tests. It's better for me than couple of months ago.
I just had an interview and they asked me prepare for coding live. I had to manually code bubble sort before the interview. It felt really nice, but it ended up them showing me some code and review it instead :facepalm:
I tell it I am a chimp with a 3 inch brain so complexity beyond chimp scale is forbidden unless requested. And to apply theory of bounded rationality during all chimp interactions. Takes care of me like a spl circus animal.
I ask it for a quick complexity thermometer reading when ever the rabbit hole is looking like it might explode.
Also walking away from it every few hours and doing random simple mindless stuff has helped. Like I will go inspect my fish and plants. Try to have a chat with the cat, get others around to do a little dance etc etc.
I have a priest in the family who has to deal with lot of heavy duty stuff/people's problems. They train them not to get totally sucked in and loose themselves. He calls it grounding. Just simple mundane things to focus shift out of anything heavy.
I am vey tired. I maintain Pinggy - my first bootstrapped startup and building Lumirank , my next one. But every day is a struggle of coping up with what is new and what 20 more competitors that corpped up yesterday are up to. Millions of lines of codes being written every day, I can not merge without knowing the overall architecture and flow. But I can't rest. Everyone is running.
Sometimes I feel please someone give me a job where I can work 12 hours a day - but the rest of the 12 hours I don't need to think about what will happen tomorrow.. It will still be less stressful.
I am surprised of how many people here are claiming they review the results of multiple agents working on different prs at the same time constantly context switching between them.
I just focus on one, I use on purpose very short interactions because I want to check every single function directly and not a whole 300lines file or 2000lines pr just at the end. Probably I am going slower than most of you but I feel I retain the knowledge of my codebase almost as good as when I was the one writing the code directly, and my PRs get very few review comments
What get me tired the most is when people submit to me a pr to review and I see immediately they full vibe coded it and just delegate to me the actual hard work of correcting the nonsense written in it with a single prompt without interactions. The worst is when you ask someone why they implemented something that way and they answer with "well calude made it, I really didn't think about it, now that I read it it makes no sense indeed"
If someone sends you vibe coded code to review, then you are obliged to give them a vibe coded review answer right back at them. Make sure your vibe review prompt tells to agent to be as nit picky as possible. Maybe tell it to be intentionally contradictory with it's change requests. Keep answering with impossible vibe coded review answers when they get back to you and nothing else.
I'm similar. I work in a context where I am responsible for the code that is in production and need to know how it works and vouch for its correctness.
Even with short interactions though, the effort in reviewing adds up and I ended up building a tool to help with it: https://github.com/imron/crt
I can review in the tool, and it remembers which files I have seen/approved since a given commit, then I can leave comments against source code and have an agent pick up those comments over mcp so it has context when making corrections.
I’ve been assigning the copilot review thing on all of these PRs, but I’ve noticed lately it’s started writing really difficult to verify review comments. It would take you literally all day to understand and test the things it’s commenting to check if they actually are issues, so as a result we all just ignore whatever the copilot review says and no one does any review.
We have removed it now, but when we used it we tweaked it to tell us only what it thoughts were major issues and ignore the rest because we had the same exact problem. Most comments were garbage
The whole situation is suboptimal because the manager uses Claude to generate tickets, the developers copy and paste the whole thing in to Claude and then post a PR they didn’t understand a thing about, and then it gets to me and I just don’t have time to review 4000 lines of code. So it ships having not had a single person understand what it does or was even for.
My entire life, I have found meaning in solving problems with code. With AI, this is no longer competitive. Now I develop more than ever and am more productive than ever, but it still doesn't feel as rewarding. I would like to go back.
There will be a generation who grow up with AI who don't know how to solve puzzles other than AI told them the answer. They will fill up the SW developer bucket quite quickly so that those who recall the pre-AI days will be vastly outnumbered. It feels bad now but it only going to get worse.
"People won't play difficult games anymore because the new Wolverine game gave them a blue line to follow everywhere and removed any need for decisions or exploration."
Yeah, I got into this field because I enjoyed solving deep technical problems mostly on my own.
Now instead I don't get to code anymore and I'm the manager of a half-dozen mid-level engineers with unflappable confidence and very, very poor long term memory.
Enforcing boundaries and saying no have gotten more important than ever, now that violation of either is automated at superhuman speeds.
FWIW, there might be a silver lining in this. A lot of dysfunction in the current state of tech can (IMO) be traced back to violation of boundaries.
This might lead to a cultural shift. Though, that might also just be wishful thinking on my part.
___
> I would love to hear your tips, but still be relevant in this AI era, so my boss does not fire me.
If your boss fires you because you didn't multiply your performance by doing more work in the same time (without your compensation changing), that's not your fault but simply your employer trying to squeeze you harder and that not working out.
Always keep that in mind in these discussions.
Which isn't to say that you should not improve efficiency and all, but it needs to be organic. If something inorganic is asked, then the (moral) failure is on the asker and not the askee.
When we get good at something, it results in flow state, but it also means that was mechanizable in a way and now AI is taking that away. So we are always grappling with the new and effortful. And the context switching, jeez. We need entirely new visualizations of work flow that recognizes we are now doing more, and in parallel. Wonder if this is how air traffic control people feel.
As someone who has retired from a programming career…
The AI era is, so far, not a problem. Obviously, my having retired is significant.
Last night I spent an hour with Claude (free tier) setting up a "modern" 6502 build environment for MacOS. By the time I went to bed I had 1) 6502 assembly syntax highlighting in VSCode, 2) was compiling 6502 assembly for the KIM-1 to generate a "paper tape" format, 3) pasting this paper tape binary into CoolTerm to send over TTY (1200-N-8-1) to my KIM-1 clone (PAL-2).
(If I'm tired it's because I excitedly stayed up later than usual to get Claude to help get this workflow working.)
On the other side of the world… a friend who is still at Apple tells me how other engineers at Apple are using Claude to make changes, create pull-requests on code he maintains with a, "Hey, can you approve this change?"
Now he has to go through the code diff that Claude generated…
Oh!, there's some stupid shit in that diff. Did the engineer who submitted the pull request even look at the code before submitting? Who knows. Either they did not or they were not qualified (or well-versed enough with the code) to spot the AI-fluff/crap.
So, yeah, that would suck—keep me busier than I would have been Antediluvian (thanks Claude, for suggesting that made-up Latin).
Sam Altman should've said only one person companies can reach a billion dollars in the future, as everyone else is too busy feeding each other awful LLM outputs.
AI-generated code works great for me, most of the time.
I'm productive, I ship, I can show stakeholders stuff in no time and and they are mostly happy.
But the development sucks.
I don't even really look at the code anymore, because reading generated code is a drag. And there is no reason really to do it.
There is no fulfillment of "I solved this real problem using some intricate logic that came out of my mind". I just throw tokens at it, and it works, great.
I don't really talk to coworkers anymore about the code. We don't do coding sessions together, or bang our heads together over some hard problem, because there are no hard problems. We don't learn from each other anymore. Everyone just asks their agents, and they are fine.
I feel my brain atrophying. I forget how to do things. Developing now feels like being kinda drunk.
Can this be solved? Absolutely: just do it like you used do! Just like quitting to smoke is easy, just don't do it, right? But what if all your colleagues go out to smoke all the time? Do you sit alone? Do you go with them, but just stand next to them in their fume? Or do you need to find new friends? Going back feels like going back to an Abacus after you invented the pocket calculator.
It’s a bit of a mixed bag for me. My brain thrives really well with keeping a lot of balls in the air at the same time and seeing quick progress and results, and I’ve always had a hard time finishing things because of the more “boring parts”, like writing documentation and proper tests etc. (yes I’m sorry for how it sounds)
But it’s not because I don’t think it’s important and that I don’t want to do it. I have a very engineering mind that likes to be thorough, and I feel like with AI I get to do all the “boring” parts, without it being boring. I get to spend more time on the architectural thinking and finding tests for edge cases, focusing on stakeholder needs and keeping the momentum by developing fast.
The biggest problem is that I’m losing grasps of the projects. I don’t review much anymore and I feel like I’m losing my ability to code.
I agree, I just mostly think of LLMs as tools that help me to complete tasks efficiently, I am not being fatigued by the AI though when it comes to projects, I let the AI write and I don't review code as much as I used to.
I'm not tired of work per se, but I'm definitely tired of what Internet is becoming: AI generated articles on AI generated websites, AI generated videos about AI generated products, AI generated emails inviting me to test AI generated features. AI generated movies and games coming soon. At work it's all AI generated presentations about how AI solves everything, AI on AI for AI with AI by AI, AI, AI, AI, AI, AI...
As agents become more capable, they can take on bigger work packages without being micromanaged. My conclusion from that is: I need to elevate myself up the hierarchy and work more like an engineering director than an engineer. For a few days, I contemplated how I want to work.
For me, the conclusion was: I want a spatial layout of work on a canvas per project and narrated demo videos, like in a feature launch review, where I as "engineering director" can decide to launch or give feedback for another iteration. Input mostly by voice. So it's a bit like sitting in a meeting room with engineers coming and going.
As software is so cheap to build nowadays, I think you just need to find out how you want to work and then build the harness for that style of work.
(If you want to try out my style, it's free and open-source. I'm sharing it because I'm excited about how much fun it is to use: https://github.com/sadilek/obeya)
I use AI to correct AI :D Yea I know, sounds ridiculous, but hear me out.
Claude Web works differently than Claude Code. I use Claude Web/Chat for briefs, general overview of the project outside of code/app, for planning and handing off the work to Code, that includes prompts. Code is tasked to output a detailed view of what it found, run measurements and tests (designed both by me and Chat) and report findings back. That is fed back to Chat and I get a rough summary. Errors are categorized by their blast-radius and so on.
Eventually I do a deep dive of what changed, how well it's handling, how much it's struggling, token usage, time spent per review/test and so on. (btw, Code is tasked to run an independent review via contextless agent on the codebase, that easily costs me another 20-30 mins and ~200K tokens on my current project, each run).
It still makes architectural issues, tries bandaiding and patching instead of properly designing, overvaluing/devaluing certain cases and misallocating where most of the focus should be, but that's relatively easier to spot and propose a fix for than doing a line by line code review.
Codebase quality is ensured through mapping the flow and the output space and having a hard-pinned tests Code can't modify to green its answers, + the independent reviewer I mentioned above.
It's not as fun as going in by hand on deep technical problems, sure. But AI usage is less about fun and more about output speed, so use it for what it is for ig? :D
I went through a period of feeling very tired. For me two things helped:
1. Being okay with being "less" productive. I'm probably still 10x more productive than I was in the pre-AI era, but in the early days of AI I would have eight Claude sessions running. Now it's two or three and I take more advantage of the free time I have between Claude assignments. My bosses are still happy with my output and I'm not being burnt out.
2. Very personal, but building my own tooling really helped. I built https://smalldocs.org / https://github.com/espressoplease/smalldocs (a document creator for AI agents) which allows me to take a lot of work outside of a chat history. E.g. I might make 10 MRs and then say "smalldoc me a list of the MRs" and I have that in my library, or if I'm considering various architectures I'll ask "smalldoc me a simple overview of the three architectures with mermaid diagrams" and then I can review the agent's thinking in a cleaner way.
I realised how easy it is for me to build the dash.
But how hard it is to sit down for 1 hour, look at the pages, scroll through them as if I am my client, notice inconsistencies, notice where the numbers don't make sense.
Sitting down, and actually checking and comparing and thinking were already hard before AI, and they became a lot harder.
I finally understand how tiring and how much work it is to manage tasks and make decisions all day.
I used to think that only the workers (the coders, burger makers) did all the hard work, while the managers and up basically sat on their butts and relaxed. But decisions and managing even without coding is just as exhausting.
Same same, it's the consequence of being "more efficient with AI" unfortunately. We've increased the amount of things we all create, but mental capacity stayed the same (because we're humans) so we have to process much more info.
What helps me from time to time is "mental offloading" - I basically sit for an hour and write down everything that bothers me, all unfinished thinkings and questions. And I'm not trying to solve them, I'm just putting them to the paper (or on my laptop / phone) to offload my mind and later I think about them in a more detailed way. I hope that helps!
The turmoil is fascinating. Everyone is uncertain.
I have been driven to reading a lot more theory and philosophy of late.
Can't recommend my latest acquisition enough: "Computational Reflections" Cantwell Smith.
The current state of affairs is stressful and at times depressing, sure. But I am very, very excited for what comes next. The insanity downward spiral that had been consuming software engineering is accelerating and being accelerated by LLMs.
Lots of people are way too distracted by money they or others are making to see this time of change for what it is: a new dawn in computing is approaching, and the fact that chatbot coding is bringing down the system that begot them is quite fitting.
I feel like we live in the ancient history of the computer age. It's like seeing writing and temples and cities first getting started and in the geologically stunningly short time of 5 thousand years we went from yet another primate to orbiting Earth. The next 500 years will see even more drastic change!
Before writing about how I feel about AI, some ideas that sometimes help me:
- plan your breaks, e.g. "1h hour of work and then go away from the computer for 5 or 10 minutes"
* try to break the "just plan a little further" loop
- accept that you cannot be concentrated for hours and that reading an AI chat is mentally challenging (see first point about breaks)
- Talk with your coworkers about it. You are not alone, it is important to talk about how you feel
- set some boundarys, e.g. I don't work in parallel on different stuff
- do NOT doom scroll while the AI is "thinking". Try some meditation or mindfullnes exercises. Doom scrolling on reddit, HN or youtube will mess up your brain even more.
- from every task, try to do something on your own, and then scale this up by AI. That tip is from a former co-worker. Their idea was that you need to figure out things first and then let AI do the boring task to scale it up. Depends on the topic sadly.
- do some things by hand, especially if they are fun for you. E.g. I really like UI design, somethign that basically is gone now for me.
It feels very strange on multiple levels.
There is the tiredness of reading text all day long. The fear of loosing my coding abilities because Claude does is. I am totally loosing track of what I did when. I can tell you about things I coded 5 years ago, I cannot tell you what I did with Claude last week.
And don't get me wrong, I LOVE planning stuff out before doing that. If there is anything positive about AI, then it is people actually PLANNING before doing.
On the other hand, I can now do/try out things I did not do because I did not have time. But at the same time, it feels like I am working on making myself not needed anymore in the future.
Worst thing, I feel like I am getting dumber. My attention spans or the "I need to know why this happens" span gets shorter, because with AI everything has to be faster?! Riiight?
I don't know how this will turn out in the future.
I see people underselling themselves becoming anxious about working without AI (either because of how they think about their abilities or management pressure).
I am getting chat snippits from AI conversations via Teams, PR checks are skills now and we need 3 rounds of discussing findings with AI ...
We live in interesting times. But yeah, I am feeling with miserable about it.
SW development skills are changing. The sheer amount and volume of AI code means eventually just accepting what it churns out and hope for the best. That's the direction it is going to converge to.
This is definitely a real thing. I think part of it is inherent to the shift, we’re now managing multiple agents which means more context switching than before.
I think part of it is just a point in time thing though (hopefully).
We can’t trust the agent output so need to check it. We can’t trust the agent to use common sense so have to think through every clarification instead of assuming a competent peer will come back with any necessary questions instead of making bad assumptions.
I think all of these things will continue to improve over time.
For myself AI quickly went from so painful to being avoided to something I can’t work without.
Indeed I am having the same mental overload, but I am starting to shift my mindset.
I realised that everyone in my company has the same mental overload.
So instead of focusing on myself, I found it easier to focus on the people I work with. What are they doing? But also scheduling, in the next meeting what will I share? Then I work backwards from that.
Have you asked your system? Like, literally, "this new way of working is mentally overstraining me. What can we do at high level, that still is compatible with me in a competitive workspace?"
Sometimes, I skim the generated text. I mean, most of it is supposition and filler, right? The rest of the time, I cut to the chase: does the code work? If not, I ignore all the generated text, pass along the errors/warnings, and go from there.
What I like best about working with AI is that I feel like a leader developer again (I've been retired from programming for almost 30 years.) It's pleasant and I don't need to worry about how snarky I get when going over bad code.
> It's pleasant and I don't need to worry about how snarky I get when going over bad code.
Yes, reviewing an LLM code is at times infuriating, but at least I can be brief and direct, rather than spend 2x the effort trying to make sure it won't be interpreted as harsh.
I use a sort of kanban board so I have a column of the backlog of things to do, a column of tasks agents are actively working on and a column of things that need review. The agents put summaries of what they did in each task and I can write my own comments there too. It’s not perfect, but it definitely helps me with context switching because I can open a task, quickly scan it to see what is going on and then open an agent with the task as its context to continue working on it.
I think it's a very exciting time and I am enthusiastic about it. What I tried to do is automate as much as possible of that decision-making, guidance and taste. Having said that, it is significantly more context switching, my solution so far is to have agents run longer autonomously, less interruptions is less switching. Prerequisite is that you must build a system first that lets you automate that, which is often very personal; most tools don't particularly apply.
Yes it's like I'm being chased by a ghost telling me I need to be more productive. And if I don't use my weekly limit to at least 90% I'm just wasting money and tokens. Should I just burn tokens endlessly? Is that a goal in itself? Where is the vision? Ami really building something worthwhile?
Some days I get far more done than my capacity and end up in an addicted state where I just can't stop working, and some days I don't even want to look at an AI chat window.. I also don't really get the feeling that I'm working.
When claude code became quite good, I became quite excited. I wanted to make sure that I consumed my full quota ( 5-hourly and weekly ). I was working on four to five different project at the same time. And it was excruciatingly painful after the initial rush wore off.
I went back to working on one project at a time to regain my sanity.
I don't get paid for my productivity increase, so I am using AI to do less.
One ticket per day, the rest of the time is used reviewing other peoples PRs,
answering emails and messages, writing tickets and so.
It’s the exact opposite for me. I don’t get tired at all from using AI. I am one of the people who enjoy building and not necessarily coding. A part of me is sad that I don’t write code anymore, I did have a thought yesterday when reviewing a PR, but working is fun for me again.
What was making me tired was all the doom and gloom online. I stopped consuming certain news and stopped going on Reddit entirely and my outlook is noticeably better.
I haven’t cut out sad news entirely but I don’t read speculative opinions about things that don’t actually affect me.
In terms of the overload I find that the clearly defined problems have much more obvious solutions. Identify your pain points and use AI to build you a tool or reusable prompt to make your life easier. It’s a constant refinement process to figure out what helps you.
For example, I was having a hard time managing multiple sessions so I had it build a window manager with notifications and color coding based on session hooks and git issue/PR status. I’m comfortably up to 8-12 sessions working in different worktrees.
If you have the ability, leave some parts of the codebase completely in your control. It's impossible to think and work deeply when using AI. So leave the most critical component to yourself and let AI handle the rest
Its crazy to see how we can now do things which were not thinkable a few years back.
Finally i read real progress i can see and use by myself and not constantly some crypto and blockchain stuff.
While I would like to have a stable world/live and for sure AI is disruptive as hell to a lot of people and potentially to me too, at least progress of technology i can't stop.
And i thought about automatistaion for decades for sure, I don't play devils advocate and thinking "yeah its great that we invented the loom" while hating AI for what it is: tech.
In worst case I only watch one 45 minute ai summary video from my YT channels, but I do have a lot of random small experiments i'm playing around with anyway.
What i discoved though: if you never go above 20$ subscriptions, you hold yourself back because you don't think about approaching bigger experiements.
I find it more difficult to get into "flow state" as the AI agents are continuously "distracting" me. Similarly to managing a team of 5 excellent engineers with poor communication boundaries.
im more wired than tired and i think a good chunk of my excitment comes from agency. ive quit my job one year ago and after a bit of wandering around discovered opus 4.5. since than i’ve been able to launch somewhat succesfull projects, switch to linux, and generally enjoy the fruits of being able to think of something and bring it to the world in a fee minutes. when i was working and “made to use ai”, i just hated it. i still remember one time the founder from previous job said “we’re going to go with what claude said”
I had something nice going with allaboutberlin.com. It helped people and covered my bills. I spent most of my tine building helpful things.
Now I'm down three quarters of my traffic and my inbox is filled with LLM-generated outreach. I'm competing with AI trained on my work, and with slop websites generated with that AI.
The original research is still necessary, but it has to happen with a dwindling audience and a shrinking revenue. I have to worry a lot more about paying rent, so I have less time to contribute to society.
Worst of all, LLM noise has filled every crevice of my life. It's billboards, blogs, music, videos, social media. It has filled the world with spam and nobody is happy, but everyone's Christmas bonus depends on adding oil onto that fire.
I mourn the loss of our internet, our culture, my career and my business all at once. And for what? We're all worse off!
I used to think that after the invention of AI, we’d become freer and wealthier. I was young and naive... Now we just work more and faster, often for less money.
My biggest problem is remembering what I did yesterday. It's starting to become hard to remember what I have going on right now. Multitasking is a killer.
Yes, the worst (or just longest) pieces of slop tire me out. When I leave work I usually cannot read or spell properly. The texts I read in the afternoon make much more sense in the next day. I'm making typos I could never ever imagine making. There's just too much to review and ignore.
I am old enough to be retired -- career spent as a research geophysicist. I have about 2-3 scientific projects/papers/codebases left in me before I go be a hermit in a cave somewhere.
LLMs appear to be able help me get through that. I'll take the help.
BUT JEEZ! Dealing with "Rain Man"[1] LLMs week-in and week-out is wildly annoying, frustrating, and anger inducing.
If they all would just STOP sounding so authoritative when spewing nonsense!
[1] After the Dustin Hoffman character in the movie of the same name. "Ah but you know I'm an excellent d̴r̴i̴v̴e̴r̴ programmer."
Not sure what harness you use, but I've found you can influence LLM "voice" a lot more by replacing system prompts rather than just adding content to a CLAUDE.md file. I see a lot of people complaining about how Claude "talks" but very little talking about how to adjust it.
The main benefits to modifying the system prompt or output style is that the harness "reminds" the model about those things every turn. So if you've experienced things like Claude Code losing track of instruction from CLAUDE.md in large contexts, it indicates you may need to add the instruction to the system prompt instead which is far more durable.
One thing you might consider, that helped me: modify your agent's personality to suit you. You may be exhausted because you're talking to someone you don't like all day. I've instructed Claude to talk to me like Confucius, because I have philosophical taste, and it makes a world of difference in terms of how I feel at the end of the day.
Claude's default personality is a bit too combative for me. Another instruction I've given it is "don't argue to win".
I've found Claude to be the least sycophantic OOTB. But I've also been running custom system prompts for a while now and have lost touch with what the default behavior looks like. I've had to balance "push back when I'm wrong" and "act like a peer" style instructions which have lead to some pretty annoying one-upmanship style dialogue.
More specifically those instructions alone caused the model to narrow in on tangentially incorrect details while completely ignoring the thrust of my argument. I get where this could cause issues with writing code, but I do a lot more than write code with CC and the conversational voice is important to me.
This is something I avoid at all costs. If it's someone else's slop I just won't read it, and if it's "my" slop I prefer to interactively ask questions about it rather than reading a wall of text.
It didn't really bother me during the altcoin/nft scam era, other than from time to time trying to explain at a normie that they were mostly scams, you know the concept that if you want to sell an asset for X you have to find somebody willing to buy.
But at least you could simply ignore all the mentally ill megalomaniacs that were like duude this is the future
This time instead it's so pervasive that you can simply not excape the narrative and simply concentrate on the good stuff about the technology. It's the narrative, the pressure, the whole global situation, not AI in general
In true drinking the kool-aid style, I've attempted to solve this problem with an LLM!
I have a skill which aggregates what I've been working on over the last day (from slack, calendar, mail etc. MCP) and writes a notion journal entry for me. It runs in the evening when I've finished and is ready the next morning when I start. It helps with the context switching - reminding me of deadlines or what I was in the middle of before I left.
More of what? I see many here using AI to build more of the same things in another language such as rewrites and then claiming that they are "productive", which is the problem of many engineers here.
Not that much risk taking if all what everyone is doing is just full blown rewrites and then abandoning it in a couple of months later and just wasting tokens for free.
If the result is to vibe code something for free and it makes no money and has no users then it is a waste of time.
Clearly there are few limits to what you can implement with AI agents today.
Depending on the complexity, you will spend a lot of time QA testing, giving feedback, and iterating.
One can take shortcuts such as porting existing software, as you mentioned. Many are just content publishing crappy novel software that they barely tested.
You don't need to do that. No one is forcing you not to QA test and iterate until you have a good product.
My boss is demanding me to use AI but hasn’t started demanding me to run parallel agents or anything like that, so I’m managing to keep things under control. Who knows how long it’ll take before they start doing that or just fire me outright. I feel pretty checked out anyway. I do remind myself every day that “it’s just a job” and I shouldn’t care about it. I think I’m more worried about AI ruining my leisure time too and has me check out of life entirely.
I'm just extremely tired of non-technical people trying to tell me how to do my job, what the state-of-the-art is, that I can implement their non-existent specs faster, or that software development is just about the ideas not implementation because we have swarms of clankers available.
I’ve had some luck in telling people I don’t need them sending me a 1000 word claudeslop technical change, I have the same model they do and can generate the same thing if needed.
But then I have to repeat this lecture to so many people it’s wearing me down. I’ve mostly resorted to just not reading it and simply ignoring any AI text sent to me.
I find the hype exhausting. Especially the pseudo intellectual essays that pop up in my Twitter feed every time I accidentally open the app.
The actual using of AI, though, doesn’t tire me much. But I mostly use it for answering specific types of questions that I’d otherwise need to spend a lot of time researching.
i didnt realize how much the mechanical act of typing on a keyboard was cognitively load-bearing for me.
that 'dead time' was when my mind was doing background processing, so when i reached another decision point, i was mentally prepared. It wasn't actually dead.
and now, working with agents, i feel much much more tired because that background processing cant happen
I don’t think it’s just fatigue, it’s more like emotional exhaustion and depression where everything seems to be going smoothly, but then, in a given moment, you realize that it’s all just complete nonsense.
> I am pretty sure most of you are experiencing the same as me, but: Have you found ways of using AI without increasing mental overload?
I would recommend you to try using FIM/line completion rather than the agent approach. Using FIM it makes me (feel) productive while still being in control. Oftentimes when writing code for production, documentation etc. the agent approach feels like a wash to me - sometimes it saves time, sometimes it wastes time. With FIM the gain/quality is more consistent.
Not personally. What I am tired of is the internet's current approach of treating every discussion (of anything) with pure outrage without considering any nuance.
It's completely destroying genuine conversations about the pros and cons of AI when people will just forever parrot that everything is slop
I am tired of the fact that even my hobbies have now been taken over by LLM-wielding imbeciles. Nothing better than going through LLM output your whole work day only to tune in to your favourite community after work to see an absolute idiot throw a slop grenade and refuse to accept that the LLM served him bullshit...
It feels a lot of reddit comments (but also here on HN) are AI generated. It would be helpful to see a kind of marking of probability a comment is generated.
The AI era just started. So no, but I would like if we weren't in such a capitalist hellscape where you practically have to eat other people to survive. The world now is grim because of who is leading. Stupid, rich sociopaths have worked really hard to make sure democracy doesn't work properly, crying about "fake news" and lying until the fourth estate is dead. We are in an era of criminal leadership more than ever. I am worried that the potential for massive species-level tragedy is building up, and we're not addressing real problems like resource scarcity and human caused mass extinctions and so on. There is a really bad future ahead when our supposed leaders say "drill baby drill", gut institutions doing positive work, go to war, and support genocide.
Yes. Everyone is forcing AI everywhere and nobody uses AI well.
AI is a net negative in terms of mental health.
It also creeps into hobbies, especially video games where it's absolutely ruined the decomp/recomp scene. AI slop recomps are a cancer on what used to be an incredible revolution in coding.
I've been enjoying it, forgive me for torturing an economics metaphor but most of the AI hype has been on the "sell" side, people can create lots of MRs and noise and potential outages. I've been working on the "buy" side building eval frameworks for all steps in our workflow so we can actually measure agent output instead of just saying "look cycle time was cut in half!" while churning out slop.
> I am pretty sure most of you are experiencing the same as me, but: Have you found ways of using AI without increasing mental overload?
Yeah. Quitting before this shit eats me alive. I'm in an EE degree program now. Better three years of study while everyone else gets to deal with the bananaware and the bullshit, and let's see then once the seas have calmed.
Until then, I actively try to avoid engaging with Slop Generator regurgitations and call it out whenever I see someone not disclosing they used a Slop Generator.
I am tired, even if I don't use AI too much for my day job designing HW, mostly for disposable scripts if at all. The utter lack of stability in the market, not even 5 years out, is sad. That is even though I have savings to weather the storm and no major liabilities.
AI is a boon for learning, but the result will heavily depend on mental capability of the user; having a depression-induced brainrot doesn't help though.
And, in the last year clankers have started coming for us EEs.
The reality is that many people are more fucked than I am, but I can't stop the bad thoughts.
> And, in the last year clankers have started coming for us EEs.
Get your practical skills in residential and commercial electricity up to speed and join the trades. The world is going for solar, heat pumps and electric vehicles no matter what corrupt politicians want (yay free market for once, lol), and that means there will be demand for people actually pulling wires and installing cabinets for decades to come. Start out as a side gig and when you find the clankers get too close to home turf, expand.
Over 40 here, and I’m having the opposite experience. Honestly, I think experience is part of why. Years of projects teach you what to ignore: which generated code actually needs a careful review, which decisions matter, and when to stop. I don’t run many things in parallel. I’d rather do one thing well than babysit five agents.
It also helps that a lot of my AI use is on my own side projects, where I set the pace. That makes it fun instead of draining. Maybe the tiredness has less to do with AI itself and more to do with being pushed to maximize throughput?
I agree with your observation. If I have kept the same targets as before, of course I would be working less. But being able to do more means being more ambitious. But the competition is also growing so in general the pattern is everybody works more
Putting all my requirements into words just to be misunderstood is the worst part. In my brain I have everything layed out. Then, the clanker will miss nuances and I have to elaborate or realgin. It would be okay if it would be a junior. But it is advertised as a superior.
I understand that faster results may be rewarding; But nothing as rewarding then refactoring some legacy application by hand and deleting 50% more LoC then adding/modifying.
Or picking the exact right abstraction now which I will use in two weeks from now and happily grind on my solid, craftsmen stones layed out beforehand.
Imho our craft was not understood beforehand. Now folks think they can grasp everything one mentions on the same level - they have "built" similiar things with ai. So it is even more frustrating to argue about stuff.
Yes, I am tired; but worse, I am unhappy.
I understand that a lot of people really enjoy managing teams of agents and are far more productive now. I am happy for them, I really am. Apparently they never actually enjoyed coding, they will show up in droves in these comments to tell you that what they really enjoyed was the end result.
I am more productive too, but I'm going to be honest - I wasn't really looking for more productivity. I am at a point in my life where boosting my productivity is absolutely not one of my personal objectives.
Using Claude Code all day is like being a STEM kid that was gifted a Lego-building-robot. What a miracle! Now you don't have to deal with all the hassle of assembling Legos - you just tell it you want a castle and boom! What a time to be alive!
yep. ive accepted it, called it a good run, put away all my fun personal projects, and started pursuing a nursing degree i can treat as a stable day job
with code all i wanted to do was code. with a nursing at least ill want to leave it at the hospital (yes nurse ptsd is real but nurse pay is great in pnw)
i think it might be good to have a job i treat as a job and not a passion i carry with me everywhere
Photography didn’t kill painting, as some people feared. A painting can be more accurate than a photo, in some ways.
I think the same will be true of programming and other skills.
No body (at least I know of) buys software for hand crafted beautiful code or collects them.
True, but a transformer system regresses to the mean. I don’t mind that with my bank’s app (in fact thats probably a good thing) but there are other kinds of applications where creativity will be required.
I really doubt we’ll have systems that come up with the Apple folding phone’s “blurry image of the outside screen as you open that snaps into focus as you finish opening it” feature, at least for a very long time. The very training system of these foundation models works against it.
Photography and painting are two entirely distinct art forms.
At the time people didn’t understand that — they thought photography would kill painting. Then for a long time photography wasn’t considered an art form. Your comment reflects an understanding/ viewpoint that is much less than a century old.
We are currently at the point equivalent to when photography was just appearing (so to speak)
I find it entirely irrelevant that contemporary comments on emerging technology of photography may have in part worried about the future of representative arts. 19th century featured a lot of blowhard know nothing know it alls.
As for today, we are discussing distinct means of producing the same thing: software. Painting remains because photography was at most displacing a subset of the field. Generative AI writing software may be displacing an entire profession.
Totally. I get value from coding tools, but I don't understand why anyone would be happy that they don't even have to look at code anymore.
It's weird times.
I feel incredibly productive. I'm building in an afternoon what would've taken weeks. I spend all my time in either MR's reviewing code, or a terminal prompting.
I've moved from being a coder, to being a product-engineer-architect. I love coding -- LOVE it. But the role has morphed. And, I actually really like the new role. I love being product centric, and caring about architecture -- it suits me wel..
The code Claude produces is often clean, but terribly archictected. So I'm spending a lot of time reviewing in detail.
But the anxienty - OMG the anxiety.
I work on multple features (generally 4-5 work streams) concurrently, so am constantly context switching. I worry about how much time I'm spending on each task. Am I doing enough code review? Too much? Too little?
What about click-through testing? Our test coverage has never been higher - should I be able to just merge these features? Wasn't that always the promise of TDD - merging with confidence? But, it feels like a risk.
And, I'm moving really really fast - but I assume my competitors are too. So that drives me to move faster. Automate more. Use claude in more places. Which brings more anxiety - am I cutting too many corners? Is this all gonna crumble on me?
Also, is everything I'm working on going to become irrelevant? Will customers buy this product, or vibe code it? Do we even have a moat? Anxiety. More anxiety.
My investors are reading stuff constantly too. They're pinging me with market updates, emerging players, shifting landscapes. More anxiety. Also - productivity tools - look - internet influencer X is now producing a gajillion lines of code a second -- are we doing that? Are we token maxing? Are we feature maxing?
Are we building the right number of internal tools? Are we automating enough? Should I evaluate the market for this new tool, or just smash something out with claude? More features. Burn more tokens. Ship ship ship.
Having to explain to investors why we're not going as fast as those guys, and building enterprise products requires care and dedication -- that takes time. Time I should be shipping, coding, featuring, tokenizing. More anxiety.
I'm exhausted.
Anyway, gotta go - herdr has pinged me four times during writing this reply. Gotta ship.
https://www.youtube.com/watch?v=_o7qjN3KF8U
Some questions that bother me:
Are people reviewing all the tests that have been created automatically?
What about existing tests? How do we make sure that AI does not fake them (by silent changes or by faking the test output)?
edit: corrected typo
>Are people reviewing all the tests that have been created automatically?
No.
> How do we make sure that AI does not fake them by silently (by silent changes or by faking the test output)?
I have never seen this happen for a couple of months at least.
Well yeah you’re not reviewing the tests so you wouldn’t catch them.
You seem to forget what tests are.
They're representative of scenarios which your application can enter.
So if the tests were mocked and a bug went live he'd still notice, after all it means your software behaved incorrectly and now you gotta clean up the pieces.
But the purpose of a test is to catch a bug before it goes live.
This is becoming cyclical, did you forget what this comment chain went about?
The discussion went from tests being written by ai, to nobody actually reviewing the tests, to the claim that given those circumstances, you wouldn't notice that the tests may not tests anything anymore because of mocks.
My statement was merely pointing out that your previous comment was groundless because they still would've found out about those bugs when they went live.
>My statement was merely pointing out that your previous comment was groundless because they still would've found out about those bugs when they went live.
Maybe. You're assuming visible and obvious bugs.
The others frequently go into prod with rigorous review too, so that hasn't really changed between pre ai and today from my perspective.
Exactly, which shows review and prod both miss things. Mutation testing (Stryker, mutmut, PIT) is how you find out which tests can’t fail.
That's a big issue in my team. We cannot really review tests manually, we already have too much just reviewing the code. And even then, sometimes the 'is this change necessary' question is forgotten or ignored (happened to me 20 minutes ago) because I an tired of fighting my junior+AI. I try to make AI refactor the tests every weeks ande that's it.
> Are people reviewing all the tests that have been created automatically?
Absolutely not! Some people ask a code assistant to write only integration/end-to-end tests and commit them. Unittests is only to help LLM generate correct code.
Yes, but periodic audits.
I'm actually unsure whether this is satire or not, which says volumes about the current moment
I feel so sorry for you and the kind of life you're building for yourself... I really hope this is a joke
I believe it.
I spend about 90% of at-computer time in a herdr-like tool.
It feels like playing factorio: whatever is manual needs to be automated away.
LLM writes the test. The test is crap. Update the skill. Ask an LLM to review your feedback on the last 5 PRs and write a skill for a review bot to look for similar problems and post. Now you only see PRs after they've gone through a few rounds of review.
The LLM has stopped picking up new issues. Ask an LLM to diagnose why. Ask an LLM to write a cron to regularly check if the LLM PR pipeline has halted and ping you. Get an LLM to try to diagnose it first.
I'm using a lot of LLMs! Ask an LLM to diagnose why. Write a tool to audit your usage, and produce a weekly report. Have it raise issue to reduce spending.
And like a base building game there are many competing facets, coming all at once.
Haha, I have been building my own software factory exactly like that. I consider it the ultimate factory game- but I sure as fuck would not want that to be my dayjob... Though I'm aware it will likely be that in a few years when the industries in my nation normalize around that.
No offense, but this feels like the most un-Factorio thing imaginable.
In factory building games, the systems in play can be understood and are fully deterministic. You're given a set of tools and a problem, and then craft together a solution. Your reward isn't just seeing the item you want on a conveyor belt, but looking at the factory you built ticking like clockwork.
What you're describing is beating out endless nondeterministic fires. It's like you were forced to play Factorio exclusively through an algorithm that designs incomprehensible factories that don't work sometimes, but designs them much faster than you ever could. So you click the magic button and see a mass of buildings appear in front of you. They may work perfectly fine, but if they don't, you need to use the same statistical factory estimator to build an in-game factory fixer mechanism and factory tester mechanism. And then a few updates later the factory fixer breaks so you need to deploy a factory fixer fixer, and a belt clean-up machine, and so on ad infinitum. To me this doesn't seem satisfying in the slightest.
I don't really see how it is as different as you say. Even if a system is stochastic, it can still be controlled, it's just a different type of engineering. Much of traditional engineering works exactly like this, with tolerances and probability rather than absolute determinism.
I think we (software engineers) are suffering from partial blindness to this aspect of engineering given that software has always been so extremely deterministic. We've trained ourselves that solving a problem means composing small fully deterministic parts into deterministic systems, and the prospect of having to work more with statistical confidence and managing failure modes feels scary and wrong.
No matter what your opinion on determinism is, the point I made that what OP is doing is very unlike a predictable and deterministic factory game still stands.
But beside that, I don't think that LLM non-determinism is really the same thing as what you see in engineering. Engineering has things like tolerances and redundancy, and while they accommodate random events, they're known, estimated and expected. They also deal with external factors, not the potential of shoddy engineering in the same system they're working on. The best way I can compare it is building a bridge that can withstand twice the normal expected full load and with enough structural redundancy to not collapse even after a prolonged lack of maintenance. These things account for many types of natural degradation and edge cases that are expected. And then there's building a bridge that you know will collapse in the next 10 years at a completely random moment, and instead of engineering a more robust solution, tacking on more gadgets to make the collapse less bad - automatic gates that close the entrances to the bridge, lake drainage system that will suck out the debris, a drone system that drops a brand new randomized bridge on the same spot, and so on. It doesn't feel very much like engineering.
If that doesn't sound convincing to you, consider one other thing. If you gave an engineer a non-deterministic solution and a deterministic solution to the exact same problem, which do you think they would prefer? Would they willingly remove determinism from a system? Maybe it's not the chaos of the real world that scares software developers so much - lots of them have to work on systems that directly interact with the real world. What if it's the fact that there is now a fight to take away their nice, understandable, satisfying deterministic solutions and inject probabilistic chaos into things that never had to have it in the first place - into absolutely everything?
> No matter what your opinion on determinism is, the point I made that what OP is doing is very unlike a predictable and deterministic factory game still stands.
I didn't mention determinism or predictability.
I said "It feels like playing factorio: whatever is manual needs to be automated away.".
I'm sure you can find many differences between Factorio and what I'm doing.
> If you gave an engineer a non-deterministic solution and a deterministic solution to the exact same problem, which do you think they would prefer? Would they willingly remove determinism from a system?
We choose non-determinism in our systems a lot, because we get better outcomes in other aspects.
* Non-deterministic network latency: Packet switching rather than circuit switching.
* Multi-tasking operating contending for the same resources.
* Multi process programs whose interleaving/execution is non-deterministic.
* Profile guided optimisation, depends on the profile.
* The entire field of randomised algorithms.
I prefer deterministic processes, but if the benefits of non-deterministic processes are significant enough, I pick non-deterministic.
Is management engineering?
Looks like written by an LLM, but it gives a realistic picture. The investors breathing down your neck is something that you don't want in this age, though.
I 100% did not write this with an LLM. I've run out of tokens.
I am fighting context switching by
- spending more time upfront working with agent on intent of the change, then design, so agent does not try to work around issues in a way that I don't intend to
- investing in ways in which agent can verify it's own change
- giving access to as much prod telemetry, etc so it can also verify the change post-merge
- obvious stuff, like automated review loops etc
But in the end I still read the "load bearing" parts of the code. Less closely, and I've long ago stopped looking at the tests. It's better for me than couple of months ago.
I just had an interview and they asked me prepare for coding live. I had to manually code bubble sort before the interview. It felt really nice, but it ended up them showing me some code and review it instead :facepalm:
Writing some code after months of agentic coding is so nice. “I remember this - it’s fun!”
>so am constantly context switching
a recipe for omni-disaster
I tell it I am a chimp with a 3 inch brain so complexity beyond chimp scale is forbidden unless requested. And to apply theory of bounded rationality during all chimp interactions. Takes care of me like a spl circus animal.
I ask it for a quick complexity thermometer reading when ever the rabbit hole is looking like it might explode.
Also walking away from it every few hours and doing random simple mindless stuff has helped. Like I will go inspect my fish and plants. Try to have a chat with the cat, get others around to do a little dance etc etc.
I have a priest in the family who has to deal with lot of heavy duty stuff/people's problems. They train them not to get totally sucked in and loose themselves. He calls it grounding. Just simple mundane things to focus shift out of anything heavy.
I am vey tired. I maintain Pinggy - my first bootstrapped startup and building Lumirank , my next one. But every day is a struggle of coping up with what is new and what 20 more competitors that corpped up yesterday are up to. Millions of lines of codes being written every day, I can not merge without knowing the overall architecture and flow. But I can't rest. Everyone is running.
Sometimes I feel please someone give me a job where I can work 12 hours a day - but the rest of the 12 hours I don't need to think about what will happen tomorrow.. It will still be less stressful.
> please someone give me a job where I can work 12 hours a day - but the rest of the 12 hours I don't need to think about what will happen tomorrow
Why do you dream about working 12h a day when 8h is pretty standard.
Just to emphasize the "but the rest of the 12 hours I don't need to think about what will happen tomorrow" part.
I am surprised of how many people here are claiming they review the results of multiple agents working on different prs at the same time constantly context switching between them.
I just focus on one, I use on purpose very short interactions because I want to check every single function directly and not a whole 300lines file or 2000lines pr just at the end. Probably I am going slower than most of you but I feel I retain the knowledge of my codebase almost as good as when I was the one writing the code directly, and my PRs get very few review comments
What get me tired the most is when people submit to me a pr to review and I see immediately they full vibe coded it and just delegate to me the actual hard work of correcting the nonsense written in it with a single prompt without interactions. The worst is when you ask someone why they implemented something that way and they answer with "well calude made it, I really didn't think about it, now that I read it it makes no sense indeed"
If someone sends you vibe coded code to review, then you are obliged to give them a vibe coded review answer right back at them. Make sure your vibe review prompt tells to agent to be as nit picky as possible. Maybe tell it to be intentionally contradictory with it's change requests. Keep answering with impossible vibe coded review answers when they get back to you and nothing else.
* insert melting face emoji here *
I'm similar. I work in a context where I am responsible for the code that is in production and need to know how it works and vouch for its correctness.
Even with short interactions though, the effort in reviewing adds up and I ended up building a tool to help with it: https://github.com/imron/crt
I can review in the tool, and it remembers which files I have seen/approved since a given commit, then I can leave comments against source code and have an agent pick up those comments over mcp so it has context when making corrections.
I’ve been assigning the copilot review thing on all of these PRs, but I’ve noticed lately it’s started writing really difficult to verify review comments. It would take you literally all day to understand and test the things it’s commenting to check if they actually are issues, so as a result we all just ignore whatever the copilot review says and no one does any review.
We have removed it now, but when we used it we tweaked it to tell us only what it thoughts were major issues and ignore the rest because we had the same exact problem. Most comments were garbage
The whole situation is suboptimal because the manager uses Claude to generate tickets, the developers copy and paste the whole thing in to Claude and then post a PR they didn’t understand a thing about, and then it gets to me and I just don’t have time to review 4000 lines of code. So it ships having not had a single person understand what it does or was even for.
The entire point of 3000 line PRs is so that you can LGTM and not feel too bad about it
My entire life, I have found meaning in solving problems with code. With AI, this is no longer competitive. Now I develop more than ever and am more productive than ever, but it still doesn't feel as rewarding. I would like to go back.
There will be a generation who grow up with AI who don't know how to solve puzzles other than AI told them the answer. They will fill up the SW developer bucket quite quickly so that those who recall the pre-AI days will be vastly outnumbered. It feels bad now but it only going to get worse.
"People won't play difficult games anymore because the new Wolverine game gave them a blue line to follow everywhere and removed any need for decisions or exploration."
Yeah, I got into this field because I enjoyed solving deep technical problems mostly on my own.
Now instead I don't get to code anymore and I'm the manager of a half-dozen mid-level engineers with unflappable confidence and very, very poor long term memory.
You and me both, man :)
Enforcing boundaries and saying no have gotten more important than ever, now that violation of either is automated at superhuman speeds.
FWIW, there might be a silver lining in this. A lot of dysfunction in the current state of tech can (IMO) be traced back to violation of boundaries. This might lead to a cultural shift. Though, that might also just be wishful thinking on my part.
___
> I would love to hear your tips, but still be relevant in this AI era, so my boss does not fire me.
If your boss fires you because you didn't multiply your performance by doing more work in the same time (without your compensation changing), that's not your fault but simply your employer trying to squeeze you harder and that not working out.
Always keep that in mind in these discussions.
Which isn't to say that you should not improve efficiency and all, but it needs to be organic. If something inorganic is asked, then the (moral) failure is on the asker and not the askee.
When we get good at something, it results in flow state, but it also means that was mechanizable in a way and now AI is taking that away. So we are always grappling with the new and effortful. And the context switching, jeez. We need entirely new visualizations of work flow that recognizes we are now doing more, and in parallel. Wonder if this is how air traffic control people feel.
As someone who has retired from a programming career…
The AI era is, so far, not a problem. Obviously, my having retired is significant.
Last night I spent an hour with Claude (free tier) setting up a "modern" 6502 build environment for MacOS. By the time I went to bed I had 1) 6502 assembly syntax highlighting in VSCode, 2) was compiling 6502 assembly for the KIM-1 to generate a "paper tape" format, 3) pasting this paper tape binary into CoolTerm to send over TTY (1200-N-8-1) to my KIM-1 clone (PAL-2).
(If I'm tired it's because I excitedly stayed up later than usual to get Claude to help get this workflow working.)
On the other side of the world… a friend who is still at Apple tells me how other engineers at Apple are using Claude to make changes, create pull-requests on code he maintains with a, "Hey, can you approve this change?"
Now he has to go through the code diff that Claude generated…
Oh!, there's some stupid shit in that diff. Did the engineer who submitted the pull request even look at the code before submitting? Who knows. Either they did not or they were not qualified (or well-versed enough with the code) to spot the AI-fluff/crap.
So, yeah, that would suck—keep me busier than I would have been Antediluvian (thanks Claude, for suggesting that made-up Latin).
> Did the engineer who submitted the pull request even look at the code before submitting? Who knows.
Of course not :D
Sam Altman should've said only one person companies can reach a billion dollars in the future, as everyone else is too busy feeding each other awful LLM outputs.
There are multiple strings coming together here.
AI-generated code works great for me, most of the time. I'm productive, I ship, I can show stakeholders stuff in no time and and they are mostly happy.
But the development sucks. I don't even really look at the code anymore, because reading generated code is a drag. And there is no reason really to do it. There is no fulfillment of "I solved this real problem using some intricate logic that came out of my mind". I just throw tokens at it, and it works, great.
I don't really talk to coworkers anymore about the code. We don't do coding sessions together, or bang our heads together over some hard problem, because there are no hard problems. We don't learn from each other anymore. Everyone just asks their agents, and they are fine.
I feel my brain atrophying. I forget how to do things. Developing now feels like being kinda drunk.
Can this be solved? Absolutely: just do it like you used do! Just like quitting to smoke is easy, just don't do it, right? But what if all your colleagues go out to smoke all the time? Do you sit alone? Do you go with them, but just stand next to them in their fume? Or do you need to find new friends? Going back feels like going back to an Abacus after you invented the pocket calculator.
It’s a bit of a mixed bag for me. My brain thrives really well with keeping a lot of balls in the air at the same time and seeing quick progress and results, and I’ve always had a hard time finishing things because of the more “boring parts”, like writing documentation and proper tests etc. (yes I’m sorry for how it sounds)
But it’s not because I don’t think it’s important and that I don’t want to do it. I have a very engineering mind that likes to be thorough, and I feel like with AI I get to do all the “boring” parts, without it being boring. I get to spend more time on the architectural thinking and finding tests for edge cases, focusing on stakeholder needs and keeping the momentum by developing fast.
The biggest problem is that I’m losing grasps of the projects. I don’t review much anymore and I feel like I’m losing my ability to code.
I agree, I just mostly think of LLMs as tools that help me to complete tasks efficiently, I am not being fatigued by the AI though when it comes to projects, I let the AI write and I don't review code as much as I used to.
I’m tired of exuberant hardware prices so that companies can shove AI down our throats on subpar products that work way worse than they did before.
This!
I'm not tired of work per se, but I'm definitely tired of what Internet is becoming: AI generated articles on AI generated websites, AI generated videos about AI generated products, AI generated emails inviting me to test AI generated features. AI generated movies and games coming soon. At work it's all AI generated presentations about how AI solves everything, AI on AI for AI with AI by AI, AI, AI, AI, AI, AI...
As agents become more capable, they can take on bigger work packages without being micromanaged. My conclusion from that is: I need to elevate myself up the hierarchy and work more like an engineering director than an engineer. For a few days, I contemplated how I want to work.
For me, the conclusion was: I want a spatial layout of work on a canvas per project and narrated demo videos, like in a feature launch review, where I as "engineering director" can decide to launch or give feedback for another iteration. Input mostly by voice. So it's a bit like sitting in a meeting room with engineers coming and going.
As software is so cheap to build nowadays, I think you just need to find out how you want to work and then build the harness for that style of work.
(If you want to try out my style, it's free and open-source. I'm sharing it because I'm excited about how much fun it is to use: https://github.com/sadilek/obeya)
I use AI to correct AI :D Yea I know, sounds ridiculous, but hear me out.
Claude Web works differently than Claude Code. I use Claude Web/Chat for briefs, general overview of the project outside of code/app, for planning and handing off the work to Code, that includes prompts. Code is tasked to output a detailed view of what it found, run measurements and tests (designed both by me and Chat) and report findings back. That is fed back to Chat and I get a rough summary. Errors are categorized by their blast-radius and so on.
Eventually I do a deep dive of what changed, how well it's handling, how much it's struggling, token usage, time spent per review/test and so on. (btw, Code is tasked to run an independent review via contextless agent on the codebase, that easily costs me another 20-30 mins and ~200K tokens on my current project, each run).
It still makes architectural issues, tries bandaiding and patching instead of properly designing, overvaluing/devaluing certain cases and misallocating where most of the focus should be, but that's relatively easier to spot and propose a fix for than doing a line by line code review.
Codebase quality is ensured through mapping the flow and the output space and having a hard-pinned tests Code can't modify to green its answers, + the independent reviewer I mentioned above.
It's not as fun as going in by hand on deep technical problems, sure. But AI usage is less about fun and more about output speed, so use it for what it is for ig? :D
I went through a period of feeling very tired. For me two things helped:
1. Being okay with being "less" productive. I'm probably still 10x more productive than I was in the pre-AI era, but in the early days of AI I would have eight Claude sessions running. Now it's two or three and I take more advantage of the free time I have between Claude assignments. My bosses are still happy with my output and I'm not being burnt out.
2. Very personal, but building my own tooling really helped. I built https://smalldocs.org / https://github.com/espressoplease/smalldocs (a document creator for AI agents) which allows me to take a lot of work outside of a chat history. E.g. I might make 10 MRs and then say "smalldoc me a list of the MRs" and I have that in my library, or if I'm considering various architectures I'll ask "smalldoc me a simple overview of the three architectures with mermaid diagrams" and then I can review the agent's thinking in a cleaner way.
I'm building a dashboard for a client.
I realised how easy it is for me to build the dash.
But how hard it is to sit down for 1 hour, look at the pages, scroll through them as if I am my client, notice inconsistencies, notice where the numbers don't make sense.
Sitting down, and actually checking and comparing and thinking were already hard before AI, and they became a lot harder.
I finally understand how tiring and how much work it is to manage tasks and make decisions all day.
I used to think that only the workers (the coders, burger makers) did all the hard work, while the managers and up basically sat on their butts and relaxed. But decisions and managing even without coding is just as exhausting.
Same same, it's the consequence of being "more efficient with AI" unfortunately. We've increased the amount of things we all create, but mental capacity stayed the same (because we're humans) so we have to process much more info.
What helps me from time to time is "mental offloading" - I basically sit for an hour and write down everything that bothers me, all unfinished thinkings and questions. And I'm not trying to solve them, I'm just putting them to the paper (or on my laptop / phone) to offload my mind and later I think about them in a more detailed way. I hope that helps!
The turmoil is fascinating. Everyone is uncertain.
I have been driven to reading a lot more theory and philosophy of late.
Can't recommend my latest acquisition enough: "Computational Reflections" Cantwell Smith.
The current state of affairs is stressful and at times depressing, sure. But I am very, very excited for what comes next. The insanity downward spiral that had been consuming software engineering is accelerating and being accelerated by LLMs.
Lots of people are way too distracted by money they or others are making to see this time of change for what it is: a new dawn in computing is approaching, and the fact that chatbot coding is bringing down the system that begot them is quite fitting.
I feel like we live in the ancient history of the computer age. It's like seeing writing and temples and cities first getting started and in the geologically stunningly short time of 5 thousand years we went from yet another primate to orbiting Earth. The next 500 years will see even more drastic change!
Before writing about how I feel about AI, some ideas that sometimes help me: - plan your breaks, e.g. "1h hour of work and then go away from the computer for 5 or 10 minutes" * try to break the "just plan a little further" loop
- accept that you cannot be concentrated for hours and that reading an AI chat is mentally challenging (see first point about breaks)
- Talk with your coworkers about it. You are not alone, it is important to talk about how you feel
- set some boundarys, e.g. I don't work in parallel on different stuff
- do NOT doom scroll while the AI is "thinking". Try some meditation or mindfullnes exercises. Doom scrolling on reddit, HN or youtube will mess up your brain even more.
- from every task, try to do something on your own, and then scale this up by AI. That tip is from a former co-worker. Their idea was that you need to figure out things first and then let AI do the boring task to scale it up. Depends on the topic sadly.
- do some things by hand, especially if they are fun for you. E.g. I really like UI design, somethign that basically is gone now for me.
It feels very strange on multiple levels.
There is the tiredness of reading text all day long. The fear of loosing my coding abilities because Claude does is. I am totally loosing track of what I did when. I can tell you about things I coded 5 years ago, I cannot tell you what I did with Claude last week.
And don't get me wrong, I LOVE planning stuff out before doing that. If there is anything positive about AI, then it is people actually PLANNING before doing.
On the other hand, I can now do/try out things I did not do because I did not have time. But at the same time, it feels like I am working on making myself not needed anymore in the future.
Worst thing, I feel like I am getting dumber. My attention spans or the "I need to know why this happens" span gets shorter, because with AI everything has to be faster?! Riiight?
I don't know how this will turn out in the future.
I see people underselling themselves becoming anxious about working without AI (either because of how they think about their abilities or management pressure).
I am getting chat snippits from AI conversations via Teams, PR checks are skills now and we need 3 rounds of discussing findings with AI ...
We live in interesting times. But yeah, I am feeling with miserable about it.
SW development skills are changing. The sheer amount and volume of AI code means eventually just accepting what it churns out and hope for the best. That's the direction it is going to converge to.
This is definitely a real thing. I think part of it is inherent to the shift, we’re now managing multiple agents which means more context switching than before.
I think part of it is just a point in time thing though (hopefully). We can’t trust the agent output so need to check it. We can’t trust the agent to use common sense so have to think through every clarification instead of assuming a competent peer will come back with any necessary questions instead of making bad assumptions. I think all of these things will continue to improve over time. For myself AI quickly went from so painful to being avoided to something I can’t work without.
Indeed I am having the same mental overload, but I am starting to shift my mindset.
I realised that everyone in my company has the same mental overload.
So instead of focusing on myself, I found it easier to focus on the people I work with. What are they doing? But also scheduling, in the next meeting what will I share? Then I work backwards from that.
Have you asked your system? Like, literally, "this new way of working is mentally overstraining me. What can we do at high level, that still is compatible with me in a competitive workspace?"
Sometimes, I skim the generated text. I mean, most of it is supposition and filler, right? The rest of the time, I cut to the chase: does the code work? If not, I ignore all the generated text, pass along the errors/warnings, and go from there. What I like best about working with AI is that I feel like a leader developer again (I've been retired from programming for almost 30 years.) It's pleasant and I don't need to worry about how snarky I get when going over bad code.
> It's pleasant and I don't need to worry about how snarky I get when going over bad code.
Yes, reviewing an LLM code is at times infuriating, but at least I can be brief and direct, rather than spend 2x the effort trying to make sure it won't be interpreted as harsh.
I use a sort of kanban board so I have a column of the backlog of things to do, a column of tasks agents are actively working on and a column of things that need review. The agents put summaries of what they did in each task and I can write my own comments there too. It’s not perfect, but it definitely helps me with context switching because I can open a task, quickly scan it to see what is going on and then open an agent with the task as its context to continue working on it.
I think it's a very exciting time and I am enthusiastic about it. What I tried to do is automate as much as possible of that decision-making, guidance and taste. Having said that, it is significantly more context switching, my solution so far is to have agents run longer autonomously, less interruptions is less switching. Prerequisite is that you must build a system first that lets you automate that, which is often very personal; most tools don't particularly apply.
Remember to take breaks, this is applicable before and after AI.
Yes it's like I'm being chased by a ghost telling me I need to be more productive. And if I don't use my weekly limit to at least 90% I'm just wasting money and tokens. Should I just burn tokens endlessly? Is that a goal in itself? Where is the vision? Ami really building something worthwhile?
Some days I get far more done than my capacity and end up in an addicted state where I just can't stop working, and some days I don't even want to look at an AI chat window.. I also don't really get the feeling that I'm working.
When claude code became quite good, I became quite excited. I wanted to make sure that I consumed my full quota ( 5-hourly and weekly ). I was working on four to five different project at the same time. And it was excruciatingly painful after the initial rush wore off.
I went back to working on one project at a time to regain my sanity.
I don't get paid for my productivity increase, so I am using AI to do less. One ticket per day, the rest of the time is used reviewing other peoples PRs, answering emails and messages, writing tickets and so.
It’s the exact opposite for me. I don’t get tired at all from using AI. I am one of the people who enjoy building and not necessarily coding. A part of me is sad that I don’t write code anymore, I did have a thought yesterday when reviewing a PR, but working is fun for me again.
What was making me tired was all the doom and gloom online. I stopped consuming certain news and stopped going on Reddit entirely and my outlook is noticeably better.
I haven’t cut out sad news entirely but I don’t read speculative opinions about things that don’t actually affect me.
In terms of the overload I find that the clearly defined problems have much more obvious solutions. Identify your pain points and use AI to build you a tool or reusable prompt to make your life easier. It’s a constant refinement process to figure out what helps you.
For example, I was having a hard time managing multiple sessions so I had it build a window manager with notifications and color coding based on session hooks and git issue/PR status. I’m comfortably up to 8-12 sessions working in different worktrees.
Refreshing take, what's bad of this AI period is all the anti-AI stuff around it.
Otherwise I've never been less stressed and chill in my working life.
I'm not sure if you could share the window manager but I'd love to get a glimpse of it, as it sounds interesting!
My job is easier than ever. Im never tired or completely drained when i get off work.
I used to grind so hard, but now it's so chill. I don't mind this change.
The main thing i dislike is how much my profession has been devalued. Code is worth so little now. It's sad.
If you have the ability, leave some parts of the codebase completely in your control. It's impossible to think and work deeply when using AI. So leave the most critical component to yourself and let AI handle the rest
I'm quite energized by this.
Its crazy to see how we can now do things which were not thinkable a few years back.
Finally i read real progress i can see and use by myself and not constantly some crypto and blockchain stuff.
While I would like to have a stable world/live and for sure AI is disruptive as hell to a lot of people and potentially to me too, at least progress of technology i can't stop.
And i thought about automatistaion for decades for sure, I don't play devils advocate and thinking "yeah its great that we invented the loom" while hating AI for what it is: tech.
In worst case I only watch one 45 minute ai summary video from my YT channels, but I do have a lot of random small experiments i'm playing around with anyway.
What i discoved though: if you never go above 20$ subscriptions, you hold yourself back because you don't think about approaching bigger experiements.
I find it more difficult to get into "flow state" as the AI agents are continuously "distracting" me. Similarly to managing a team of 5 excellent engineers with poor communication boundaries.
I've been leaning towards investing more time writing, to balance all the random reading.
Also, trying to put more and more focus time, in which I don't allow myself to use AI.
I find it very rewarding.
im more wired than tired and i think a good chunk of my excitment comes from agency. ive quit my job one year ago and after a bit of wandering around discovered opus 4.5. since than i’ve been able to launch somewhat succesfull projects, switch to linux, and generally enjoy the fruits of being able to think of something and bring it to the world in a fee minutes. when i was working and “made to use ai”, i just hated it. i still remember one time the founder from previous job said “we’re going to go with what claude said”
I don't no, but I can understand why others might.
For me it's very exciting and AI has been so far a crazy enabler for me. Both in my work and my personal life.
But I also enjoy keeping up with it and reading about it etc.
I had something nice going with allaboutberlin.com. It helped people and covered my bills. I spent most of my tine building helpful things.
Now I'm down three quarters of my traffic and my inbox is filled with LLM-generated outreach. I'm competing with AI trained on my work, and with slop websites generated with that AI.
The original research is still necessary, but it has to happen with a dwindling audience and a shrinking revenue. I have to worry a lot more about paying rent, so I have less time to contribute to society.
Worst of all, LLM noise has filled every crevice of my life. It's billboards, blogs, music, videos, social media. It has filled the world with spam and nobody is happy, but everyone's Christmas bonus depends on adding oil onto that fire.
I mourn the loss of our internet, our culture, my career and my business all at once. And for what? We're all worse off!
I used to think that after the invention of AI, we’d become freer and wealthier. I was young and naive... Now we just work more and faster, often for less money.
Capital is becoming freer from labor worries and wealthier.
My biggest problem is remembering what I did yesterday. It's starting to become hard to remember what I have going on right now. Multitasking is a killer.
As most of other comments mentioned, I am so productive and doing five side projects at the same time but I am sad.
Yes, the worst (or just longest) pieces of slop tire me out. When I leave work I usually cannot read or spell properly. The texts I read in the afternoon make much more sense in the next day. I'm making typos I could never ever imagine making. There's just too much to review and ignore.
I am old enough to be retired -- career spent as a research geophysicist. I have about 2-3 scientific projects/papers/codebases left in me before I go be a hermit in a cave somewhere.
LLMs appear to be able help me get through that. I'll take the help.
BUT JEEZ! Dealing with "Rain Man"[1] LLMs week-in and week-out is wildly annoying, frustrating, and anger inducing.
If they all would just STOP sounding so authoritative when spewing nonsense!
[1] After the Dustin Hoffman character in the movie of the same name. "Ah but you know I'm an excellent d̴r̴i̴v̴e̴r̴ programmer."
Not sure what harness you use, but I've found you can influence LLM "voice" a lot more by replacing system prompts rather than just adding content to a CLAUDE.md file. I see a lot of people complaining about how Claude "talks" but very little talking about how to adjust it.
https://code.claude.com/docs/en/agent-sdk/modifying-system-p...
The main benefits to modifying the system prompt or output style is that the harness "reminds" the model about those things every turn. So if you've experienced things like Claude Code losing track of instruction from CLAUDE.md in large contexts, it indicates you may need to add the instruction to the system prompt instead which is far more durable.
No, infact i'm more empowered than ever.. might burn out tho.. we'll see
One thing you might consider, that helped me: modify your agent's personality to suit you. You may be exhausted because you're talking to someone you don't like all day. I've instructed Claude to talk to me like Confucius, because I have philosophical taste, and it makes a world of difference in terms of how I feel at the end of the day.
Claude's default personality is a bit too combative for me. Another instruction I've given it is "don't argue to win".
I don't use Claude but this is the first time I hear someone say it's combative. People usually joke about it being too sycophantic.
I've found Claude to be the least sycophantic OOTB. But I've also been running custom system prompts for a while now and have lost touch with what the default behavior looks like. I've had to balance "push back when I'm wrong" and "act like a peer" style instructions which have lead to some pretty annoying one-upmanship style dialogue.
More specifically those instructions alone caused the model to narrow in on tangentially incorrect details while completely ignoring the thrust of my argument. I get where this could cause issues with writing code, but I do a lot more than write code with CC and the conversational voice is important to me.
> We need to keep track of many things happening in parallel...
That your workplace, not "AI era".
Fair point
I feel tired watching AI generate my code and doom scrolling on instagram like they should make the code generated within minutes
> read enormous amounts of generated text
This is something I avoid at all costs. If it's someone else's slop I just won't read it, and if it's "my" slop I prefer to interactively ask questions about it rather than reading a wall of text.
I'm just so tired of the narrative.
It didn't really bother me during the altcoin/nft scam era, other than from time to time trying to explain at a normie that they were mostly scams, you know the concept that if you want to sell an asset for X you have to find somebody willing to buy.
But at least you could simply ignore all the mentally ill megalomaniacs that were like duude this is the future
This time instead it's so pervasive that you can simply not excape the narrative and simply concentrate on the good stuff about the technology. It's the narrative, the pressure, the whole global situation, not AI in general
Yes, feeling this 100%
In true drinking the kool-aid style, I've attempted to solve this problem with an LLM!
I have a skill which aggregates what I've been working on over the last day (from slack, calendar, mail etc. MCP) and writes a notion journal entry for me. It runs in the evening when I've finished and is ready the next morning when I start. It helps with the context switching - reminding me of deadlines or what I was in the middle of before I left.
> Although AI lets us do much more,
More of what? I see many here using AI to build more of the same things in another language such as rewrites and then claiming that they are "productive", which is the problem of many engineers here.
Not that much risk taking if all what everyone is doing is just full blown rewrites and then abandoning it in a couple of months later and just wasting tokens for free.
If the result is to vibe code something for free and it makes no money and has no users then it is a waste of time.
More of whatever you had in mind.
Clearly there are few limits to what you can implement with AI agents today.
Depending on the complexity, you will spend a lot of time QA testing, giving feedback, and iterating.
One can take shortcuts such as porting existing software, as you mentioned. Many are just content publishing crappy novel software that they barely tested.
You don't need to do that. No one is forcing you not to QA test and iterate until you have a good product.
My boss is demanding me to use AI but hasn’t started demanding me to run parallel agents or anything like that, so I’m managing to keep things under control. Who knows how long it’ll take before they start doing that or just fire me outright. I feel pretty checked out anyway. I do remind myself every day that “it’s just a job” and I shouldn’t care about it. I think I’m more worried about AI ruining my leisure time too and has me check out of life entirely.
instead of working 4-5 days per week i work 1-2 intense ones.
rest i use to think
I'm just extremely tired of non-technical people trying to tell me how to do my job, what the state-of-the-art is, that I can implement their non-existent specs faster, or that software development is just about the ideas not implementation because we have swarms of clankers available.
I’ve had some luck in telling people I don’t need them sending me a 1000 word claudeslop technical change, I have the same model they do and can generate the same thing if needed.
But then I have to repeat this lecture to so many people it’s wearing me down. I’ve mostly resorted to just not reading it and simply ignoring any AI text sent to me.
I find the hype exhausting. Especially the pseudo intellectual essays that pop up in my Twitter feed every time I accidentally open the app.
The actual using of AI, though, doesn’t tire me much. But I mostly use it for answering specific types of questions that I’d otherwise need to spend a lot of time researching.
> so my boss does not fire me
That't the key. Not AI.
i didnt realize how much the mechanical act of typing on a keyboard was cognitively load-bearing for me.
that 'dead time' was when my mind was doing background processing, so when i reached another decision point, i was mentally prepared. It wasn't actually dead.
and now, working with agents, i feel much much more tired because that background processing cant happen
I don’t think it’s just fatigue, it’s more like emotional exhaustion and depression where everything seems to be going smoothly, but then, in a given moment, you realize that it’s all just complete nonsense.
> I am pretty sure most of you are experiencing the same as me, but: Have you found ways of using AI without increasing mental overload?
I would recommend you to try using FIM/line completion rather than the agent approach. Using FIM it makes me (feel) productive while still being in control. Oftentimes when writing code for production, documentation etc. the agent approach feels like a wash to me - sometimes it saves time, sometimes it wastes time. With FIM the gain/quality is more consistent.
I hate it and I know it'll get worse
Not personally. What I am tired of is the internet's current approach of treating every discussion (of anything) with pure outrage without considering any nuance.
It's completely destroying genuine conversations about the pros and cons of AI when people will just forever parrot that everything is slop
I am tired of the fact that even my hobbies have now been taken over by LLM-wielding imbeciles. Nothing better than going through LLM output your whole work day only to tune in to your favourite community after work to see an absolute idiot throw a slop grenade and refuse to accept that the LLM served him bullshit...
It feels a lot of reddit comments (but also here on HN) are AI generated. It would be helpful to see a kind of marking of probability a comment is generated.
Reddit as a platform is completely dead at this point. It’s nothing like it once was.
The AI era just started. So no, but I would like if we weren't in such a capitalist hellscape where you practically have to eat other people to survive. The world now is grim because of who is leading. Stupid, rich sociopaths have worked really hard to make sure democracy doesn't work properly, crying about "fake news" and lying until the fourth estate is dead. We are in an era of criminal leadership more than ever. I am worried that the potential for massive species-level tragedy is building up, and we're not addressing real problems like resource scarcity and human caused mass extinctions and so on. There is a really bad future ahead when our supposed leaders say "drill baby drill", gut institutions doing positive work, go to war, and support genocide.
Yes. Everyone is forcing AI everywhere and nobody uses AI well.
AI is a net negative in terms of mental health.
It also creeps into hobbies, especially video games where it's absolutely ruined the decomp/recomp scene. AI slop recomps are a cancer on what used to be an incredible revolution in coding.
I've been enjoying it, forgive me for torturing an economics metaphor but most of the AI hype has been on the "sell" side, people can create lots of MRs and noise and potential outages. I've been working on the "buy" side building eval frameworks for all steps in our workflow so we can actually measure agent output instead of just saying "look cycle time was cut in half!" while churning out slop.
> I am pretty sure most of you are experiencing the same as me, but: Have you found ways of using AI without increasing mental overload?
Yeah. Quitting before this shit eats me alive. I'm in an EE degree program now. Better three years of study while everyone else gets to deal with the bananaware and the bullshit, and let's see then once the seas have calmed.
Until then, I actively try to avoid engaging with Slop Generator regurgitations and call it out whenever I see someone not disclosing they used a Slop Generator.
I am tired, even if I don't use AI too much for my day job designing HW, mostly for disposable scripts if at all. The utter lack of stability in the market, not even 5 years out, is sad. That is even though I have savings to weather the storm and no major liabilities.
AI is a boon for learning, but the result will heavily depend on mental capability of the user; having a depression-induced brainrot doesn't help though.
And, in the last year clankers have started coming for us EEs.
The reality is that many people are more fucked than I am, but I can't stop the bad thoughts.
> And, in the last year clankers have started coming for us EEs.
Get your practical skills in residential and commercial electricity up to speed and join the trades. The world is going for solar, heat pumps and electric vehicles no matter what corrupt politicians want (yay free market for once, lol), and that means there will be demand for people actually pulling wires and installing cabinets for decades to come. Start out as a side gig and when you find the clankers get too close to home turf, expand.
Over 40 here, and I’m having the opposite experience. Honestly, I think experience is part of why. Years of projects teach you what to ignore: which generated code actually needs a careful review, which decisions matter, and when to stop. I don’t run many things in parallel. I’d rather do one thing well than babysit five agents.
It also helps that a lot of my AI use is on my own side projects, where I set the pace. That makes it fun instead of draining. Maybe the tiredness has less to do with AI itself and more to do with being pushed to maximize throughput?
I agree with your observation. If I have kept the same targets as before, of course I would be working less. But being able to do more means being more ambitious. But the competition is also growing so in general the pattern is everybody works more
Putting all my requirements into words just to be misunderstood is the worst part. In my brain I have everything layed out. Then, the clanker will miss nuances and I have to elaborate or realgin. It would be okay if it would be a junior. But it is advertised as a superior.
I understand that faster results may be rewarding; But nothing as rewarding then refactoring some legacy application by hand and deleting 50% more LoC then adding/modifying. Or picking the exact right abstraction now which I will use in two weeks from now and happily grind on my solid, craftsmen stones layed out beforehand.
Imho our craft was not understood beforehand. Now folks think they can grasp everything one mentions on the same level - they have "built" similiar things with ai. So it is even more frustrating to argue about stuff.
It is the worst technology ever emerged imho.
No, love AI. Greatest thing that ever happened. Very tired of the haters though, they ruin everything.