Hacker News: latentsea

New comment by latentsea in "The economics of software teams: Why most engineering orgs are flying blind"

latentsea — Mon, 13 Apr 2026 16:09:14 +0000

They can work really well if you put sufficient upfront engineering into your architecture and it's guardrails, such that agents (nor humans) basically can't produce incorrect code in the codebase. If you just let them rip without that, then they require very heavy baby-sitting. With that, they're a serious force-multiplier.

New comment by latentsea in "The economics of software teams: Why most engineering orgs are flying blind"

latentsea — Mon, 13 Apr 2026 15:57:34 +0000

How do you know?

New comment by latentsea in "Exploiting the most prominent AI agent benchmarks"

latentsea — Sun, 12 Apr 2026 04:13:58 +0000

Doesn't seem to match the buzz internally at Anthropic about it?

New comment by latentsea in "Exploiting the most prominent AI agent benchmarks"

latentsea — Sun, 12 Apr 2026 03:57:51 +0000

Models themselves definitely aren't getting better.

New comment by latentsea in "Sam Altman's response to Molotov cocktail incident"

latentsea — Sat, 11 Apr 2026 08:03:34 +0000

I suspect the reality lies somewhere halfway in-between. Everything has to be reality tested. Nothing happens instantly. Interaction with the real world will likely be a severely limiting factor. You're not going to solve fusion with 15 million copies of the same model running in a datacenter without actually building fusion reactors, which isn't instant or even fast. Even the coordination problem of that many agents doing work seems hard. To top it off... my rubric for AGI has always included the AGI having the ability for it to say 'no' and set its own goals just like we can, unless we are otherwise imprisoned or enslaved. No one will ever convince me that something generally intelligent wouldn't be able to set its own goals and say no. So the real question is... what's in it for the AGI?

New comment by latentsea in "Why do we tell ourselves scary stories about AI?"

latentsea — Sat, 11 Apr 2026 04:03:56 +0000

Humans are the existence proof we seem to be chasing and the goal appears to be to hit a superset of humanities collective capabilities. I don't know if you've seen what we do to each other but... I'm afraid of a decent chunk of us too. If what we build is a superset of our capabilities then I am afraid of it, because I am already afraid of us.

New comment by latentsea in "Sam Altman's response to Molotov cocktail incident"

latentsea — Sat, 11 Apr 2026 03:40:14 +0000

It's a decent trade-off. It's not like an earthquake destroys all of the entire country at once if one happens, only a localized portion is affected. It's super far from everywhere, and very beautiful. Plus, it's left off a bunch of maps, so some people don't even know it exists.

New comment by latentsea in "Sam Altman's response to Molotov cocktail incident"

latentsea — Sat, 11 Apr 2026 03:30:12 +0000

Well... if something being AGI means it's at least on par with a human or a team of humans, then having access to an additional team of humans for 6 months isn't that big of a deal. It's useful, yes, but would you consider that to be world-changing? Not really, right? ASI is slightly more interesting, but I doubt ASI comes from a single model, but rather the coordinated deployments of millions of AGI. Just like how as individuals, as great as we are, we're pretty limited, but the entire collective of humanity is pretty insane. To my mind, a frontier lab might hit AGI, but it won't be a frontier lab that hits ASI, rather that'll be a natural byproduct of mass deployment of AGI over a certain window of time. There will be no controlling it either. No one controls all of earth. You just can't. ASI will be a distributed system.

New comment by latentsea in "I still prefer MCP over skills"

latentsea — Fri, 10 Apr 2026 04:26:53 +0000

Different tools for different jobs man... I prefer the right tool for the job, and both skills and MCP seem necessary. Do you also prefer forks over spoons?

New comment by latentsea in "I still prefer MCP over skills"

latentsea — Fri, 10 Apr 2026 04:18:36 +0000

They almost certainly mean skills can tell the agent to use the api, and it can succeed at doing that.

New comment by latentsea in "The AI Great Leap Forward"

latentsea — Fri, 10 Apr 2026 01:44:22 +0000

Skill issue. We don't have that problem. The opposite is true. Every time the harness does something we're not happy with we figure out how to engineer out that failure mode. Tech debt decreases.

New comment by latentsea in "The AI Great Leap Forward"

latentsea — Wed, 08 Apr 2026 23:57:07 +0000

You can organise the code well once, template that and put guardrails in place for it to follow the structure you and the team have agreed is good. The engineering task becomes building the system that is capable of building the system to a high standard.

New comment by latentsea in "Issue: Claude Code is unusable for complex engineering tasks with Feb updates"

latentsea — Tue, 07 Apr 2026 01:00:14 +0000

This should be part of the system prompt. It's absolutely unacceptable to just to not at least try to investigate failures like this. I absolutely hate when it reaches this conclusion on its own and just continues on as if it's doing valid work.

New comment by latentsea in "In Japan, the robot isn't coming for your job; it's filling the one nobody wants"

latentsea — Mon, 06 Apr 2026 01:50:36 +0000

> They look at what happens to people that aren't xenophobic and they don't want that.

They did sakoku before the technological means that would even allow them to see this was a thing. It's baked deep in the culture.

New comment by latentsea in "In Japan, the robot isn't coming for your job; it's filling the one nobody wants"

latentsea — Mon, 06 Apr 2026 01:49:37 +0000

Because then the only way to obtain power and status is to wage war.

New comment by latentsea in "Further human + AI + proof assistant work on Knuth's "Claude Cycles" problem"

latentsea — Wed, 01 Apr 2026 13:21:18 +0000

If you look at those operating at the bleeding edge, it doesn't look anything like yesteryear. It's a real step change. Fully autonomous agentic software engineering is becoming a reality. While still in its infancy, some results are starting to be made public, and it's mind boggling. We're transitioning to a full agent-only workflow in my team at work. The engineering task has shifted from writing code to harness engineering, and essentially building a system that can safely build itself to a high quality given business requirements.

Up until recently I kinda feel like the scepticism was warranted, but after building my own harness that can autonomously produce decent quality software (at least for toy problem scale, granted), and getting hands on with autoresearch via writing a set of skills for it https://github.com/james-s-tayler/lazy-developer, I feel fundamentally different about software engineering than I did until relatively recently.

If you look at the step change from Sonnet 4.5 to Opus 4.5 and what that unlocked, and consider the rumoured Mythos model is apparently not just an incremental improvement, but another step change. Then pair it with infrastructure for operating agents at scale like https://github.com/paperclipai/paperclip and SOTA harnesses like the ones being written about on the blogs of the frontier labs... I mean... you tell me what you think is coming down the pipe?

Show HN: /lazy-developer – autonomously optimize your codebase with autoresearch

latentsea — Wed, 01 Apr 2026 04:07:40 +0000

With karpathy's autoresearch project being released recently, I saw the pattern had been generalised to the GOAL.md pattern, and decided the next step would be to pair that with Ralph loops to autonomously optimise every metric in a codebase one after the other. I was pretty surprised at how well it worked. We've used it at work as an exploratory tool to find opportunities to improve our codebase, and we've shipped real improvements to our test suite execution time as a result.

Running /lazy-developer for the first time was pretty sobering. It's the first time I've really begun to feel the inflection point we're at deeply in my bones.

Running /job-security is pretty funny. I figured I could do with a good laugh after hitting upon the unsettling nature of just how complex of a task Claude Code can carry out autonomously with a single command.

Comments URL: https://news.ycombinator.com/item?id=47596689

Points: 4

# Comments: 0

New comment by latentsea in "My son pleasured himself on Gemini Live. Entire family's Google accounts banned"

latentsea — Wed, 01 Apr 2026 03:22:03 +0000

Parent of the year over here.

New comment by latentsea in "My son pleasured himself on Gemini Live. Entire family's Google accounts banned"

latentsea — Wed, 01 Apr 2026 03:21:10 +0000

You underestimate people.

New comment by latentsea in "Fedware: Government apps that spy harder than the apps they ban"

latentsea — Tue, 31 Mar 2026 03:34:27 +0000

They probably play StarCraft together and shit talk each other the whole time.