ramraj07

Its outrageous that no one pointed out git. Its the perfect example of this. I consider code version control a solved problem post git.

Apparently the phrase cargo cult software engineering is not common anymore. Explains these things perfectly.

Yep, spend 100 billion on what should have cost 1/50that cost, and send people up to the moon with rockets that we are still keeping our fingers crossed wont kill them tomorrow, and we have to congratulate them for dunking on some irrelevant career?

The problem they solved isn't easy. But its not some insane technical breakthrough either. Literally add redundancy, thats the ask. They didnt invent quantum computing to solve the issue did they? Why dunk on sprints?

It got them to where they need to be to then worry about ECC. This is like the dudes who deploy their blog on kubernetes just in case it hits front page of new york times or something.

I take the opposite message from that line - out of touch teams working on something so over budget and so overdue, and so bureaucratic, and with such an insanely poor history of success, and they talk as if they have cured cancer.

This is the equivalent of Altavista touting how amazing their custom server racks are when Google just starts up on a rack of naked motherboards and eats their lunch and then the world.

Lets at least wait till the capsule comes back safely before touting how much better they are than "DevOps" teams running websites, apparently a comparison that's somehow relevant here to stoke egos.

As someone who's been trying to figure out the same problems in a marginally different situation, it is breathtaking to see the number of architectural decisions they have made and have also decided to discuss publicly.

I would still not use this service, personally, to build my agents, because we are then stuck with anthropic for good. While I do bet on them being the best at it for the near future, this seems like a bad abstraction layer strategically for anyone wanting to build a business building and running agents.

I can anticipate this starting to happen: shoot into the sky, wait for flock drone to start coming in, then shoot it! Free target practice!

To rich people, the privacy attack isn't an issue. We already track their private jets, how is this any different?

Im a libertarian at heart so cant argue too much with people's personal choices, but I can only say that given what I know can go wrong in even the best maintained lab, I wouldn't ever do what youre doing. Yes its all great today, but the day something goes wrong, you may not be happy finding out how its affecting you. Look up the bayer factor viii hiv incident.

Thats the point. If even such a sycophantic ai disagrees with your points you have a problem

So you agree that this whole topic is in a sliding scale of lunacy, youre just on the saner side a bit?

Test what, exactly? Purity? LPS contamination? They cant test for every last picogram of material in it. Did they test for viral contamination?

Even drug addicts heat up the thing they inject so theyre actually safer than you can ever be. Dont inject things from China into your blood!

Im sorry for your quality of life problems but calling doctors bad at statistics and then giving anecdotal evidence as proof has to start ringing some logical bells right? You dont even have to take our word. Use an LLM as judge. Paste your comment into chatgpt and see what it says.

You're applying US law. I dont think this passes muster on all Indian laws.

The advantage is marginal and sporadic, and its slower. Why would I use it over anthropic?

We have done SOC2 and it's not fake. Its real and enforced some good practices and we spent a lot of time collecting evidence and submitting it. You can take it seriously or you can choose not to.

give people the ability to curate their own collections and publish them

If you think this is not a competitive advantage then youre missing the point. LLMs arent so good that they work through bad abstractions and pretty much everyone has bad abstractions. CC is what invents some of the best abstractions (not the first). I think theyre they first ones who nailed subagents well. Theres a lot to learn from them and while im learning a lot from their source code my heart bleeds that this happened to them.

Sincerely, someone running a team building similar things for analytics.

What exactly is shitty here? A program i use for hours every day to do the job previously done by many N human beings, without many bugs, seems to have code thats seemingly messy but still clearly works.

Browse through codex and think if anyone cares about the quality of the code before Open sourcing it.

You are still responsible for the product; the code has stopped being what defines the product.

I know at least one person who joined a Michigan startup, moved over, got sued by non compete, and the new employer just didnt want any hassle and laid them off. This person had to leave country then.

The take home is dont take tech jobs in states where non-compete clauses are still legal.

Start in a tmux session and let it run ?

What country is it? Surprised someone on tech even asks this question

Employees, especially ones as well leveraged and overpaid as software engineers, are not victims. They can leave. They _should_ leave. Great engineers are still able to bet better paying jobs all the time.

This exactly matches our findings: if we start molding the repo to be "AI native" whatever that means, add the right tooling and still demand all engineers take full responsibility for their output, this system is a true multiplier.

I also have Copilot and Cursor bugbot reviews and run it on a Ralph wiggum loop with claude code. A few rounds overnight and the PR is perfect and ready for a final review before merging.

I do run 4 CC sessions in parallel though, but thats just one day a week. The rest of the week is spent figuring out the next set of features and fixes needed, operational things, meetings,feedback, etc.

Why did your company get claude code with token billing instead of getting everyone max plans ?

I used to think that the people who keep saying (in March 2026) that AI does not generate good code are just not smart and ask stupid prompts.

I think I've amended that thought. They are not necessarily lacking in intelligence. I hypothesize that LLMs pick up on optimism and pessimism among other sentiments in the incoming prompt: someone prompting with no hope that the result will be useful end up with useless garbage output and vice versa.

If you dont take a stand and refuse to clean their mess, aren't you part of the problem? No self respecting proponent of AI enabled development should suggest that the engineers generating the code are still not personally responsible for its quality.