Hacker News: rco8786

New comment by rco8786 in "Constraint Decay: The Fragility of LLM Agents in Back End Code Generation"

rco8786 — Mon, 25 May 2026 12:33:13 +0000

Well yes, ideally. But real world codebases aren't clean enough to be used as the example ideal. Styles change over time, there are always code migrations and refactors in flight, legacy code exists, etc. Using specific examples of what you expect the LLM (and humans) to do now is necessary.

New comment by rco8786 in "Claude is not your architect. Stop letting it pretend"

rco8786 — Sun, 24 May 2026 21:09:34 +0000

> AI is only as good as the person using it

Nailed it.

That said, it couldn't have possibly been that bad if you got it to "acceptable" in 2 weeks.

New comment by rco8786 in "Open source Kanban desktop app that runs parallel agents on every card"

rco8786 — Sat, 23 May 2026 13:45:42 +0000

This is fine for one off tools and I do the same. But building long-lived "professional grade" production software this fails real quickly.

My team is using AI for most of the code, but the human review layer is crucial and unavoidable if you're interested in things like reliability, uptime, controlled feature rollouts, the integrity if your user's data, etc.

New comment by rco8786 in "Project Glasswing: An Initial Update"

rco8786 — Sat, 23 May 2026 11:13:33 +0000

Software engineers generate security bugs, Software engineers find them, then Software engineers generate fix, collect salary, profit?

New comment by rco8786 in "Rubish: A Unix shell written in pure Ruby"

rco8786 — Sat, 23 May 2026 11:09:54 +0000

Yea. Modern Ruby is "fast enough", but it's very real that when Ruby was hitting its peak it was dog slow. It's hard to shake those sorts of reputations (similar to the "can't scale" reputation that Rails got because of Twitter)

New comment by rco8786 in "MIT: 20% drop in incoming graduate students"

rco8786 — Thu, 14 May 2026 15:04:21 +0000

Nothing to do with AI here, it's about immigration.

New comment by rco8786 in "MacBook Neo Deep Dive: Benchmarks, Wafer Economics, and the 8GB Gamble"

rco8786 — Wed, 13 May 2026 22:11:42 +0000

I feel confident in saying that I am better at computers than 99.99% of the general population and I have no clue what “SS” or blue USB ports are supposed to indicate.

New comment by rco8786 in "Cloudflare to cut about 20% of its workforce"

rco8786 — Fri, 08 May 2026 17:46:10 +0000

In the US??

New comment by rco8786 in "Agents can now create Cloudflare accounts, buy domains, and deploy"

rco8786 — Wed, 06 May 2026 10:24:21 +0000

> I am also not sure who Stripe Atlas for.

This was such a weird mention to see in the article. Stripe Atlas is a service that helps new businesses incorporate and onboard onto Stripe/partner services with some startup credits. It's been around forever, has nothing to do with AI, and is generally a very well-respected service.

New comment by rco8786 in "Learn Claude Code by doing, not reading"

rco8786 — Thu, 02 Apr 2026 11:41:41 +0000

Ok? I'm just explaining what claude code is, not pontificating about the capabilities of AI.

New comment by rco8786 in "Learn Claude Code by doing, not reading"

rco8786 — Mon, 30 Mar 2026 21:57:12 +0000

Claude Code is a tool that uses natural language ai systems. It itself is not a natural language ai system.

New comment by rco8786 in "AI overly affirms users asking for personal advice"

rco8786 — Sat, 28 Mar 2026 17:16:24 +0000

If they’re reaching the same results across a variety of the most popular public models, it doesn’t seem like that big a deal to know if it was Opus 4 or Opus 4.5

New comment by rco8786 in "Taming LLMs: Using Executable Oracles to Prevent Bad Code"

rco8786 — Thu, 26 Mar 2026 23:54:30 +0000

Yea this feels like saying “if you give them good enough specs they’ll produce the code you want” which reduces to…writing the code yourself. Just with more steps.

New comment by rco8786 in "GitHub is once again down"

rco8786 — Tue, 24 Mar 2026 23:40:21 +0000

GH is ripe for a disruptor right now.

New comment by rco8786 in "If DSPy is so great, why isn't anyone using it?"

rco8786 — Mon, 23 Mar 2026 16:20:47 +0000

I think they're just saying that data extraction tasks are easy to evaluate because for a given input text/file you can specify the exact structured output you expect from it.

New comment by rco8786 in "Node.js worker threads are problematic, but they work great for us"

rco8786 — Sun, 22 Mar 2026 14:34:08 +0000

Yea, for the vast, vast majority of workloads just forking separate node process ends up being better than mucking with threads.

New comment by rco8786 in "OpenCode – Open source AI coding agent"

rco8786 — Sat, 21 Mar 2026 12:29:31 +0000

> they add, remove, refine, change, fix, and break features constantly at that accelerated pace.

I wonder how much of this is because the maintainers are using OpenCode to vibe the code for OpenCode.

New comment by rco8786 in "Our commitment to Windows quality"

rco8786 — Fri, 20 Mar 2026 20:09:42 +0000

> More taskbar customization, including vertical and top positions:

When did they get rid of that?

New comment by rco8786 in "Rob Pike’s Rules of Programming (1989)"

rco8786 — Wed, 18 Mar 2026 15:04:28 +0000

Ditto here

New comment by rco8786 in "Hammerspoon"

rco8786 — Sat, 14 Mar 2026 00:58:08 +0000

Love Hammerspoon