Hacker News: nkko

New comment by nkko in "GLM-5.1: Towards Long-Horizon Tasks"

nkko — Tue, 07 Apr 2026 18:45:17 +0000

Yes, this is frustrating, but it doesn’t occur in CC. I run the conversation logs through an agent and opencode source, and it identified an issue in the reasoning implementation of opencode for Zai models. Consequently, I ceased my research and opted to use CC instead.

Durable Researcher

nkko — Tue, 07 Apr 2026 18:29:37 +0000

Article URL: https://github.com/steel-experiments/durable-researcher

Comments URL: https://news.ycombinator.com/item?id=47679402

Points: 1

# Comments: 0

The Hard Problems Nobody Has Solved

nkko — Tue, 07 Apr 2026 16:01:43 +0000

Article URL: https://www.nibzard.com/hard-problems

Comments URL: https://news.ycombinator.com/item?id=47677378

Points: 2

# Comments: 0

Absurd: Simplest durable execution workflow system

nkko — Mon, 06 Apr 2026 18:55:08 +0000

Article URL: https://github.com/earendil-works/absurd

Comments URL: https://news.ycombinator.com/item?id=47665221

Points: 3

# Comments: 0

New comment by nkko in "The CMS is dead, long live the CMS"

nkko — Sun, 05 Apr 2026 13:12:06 +0000

What happened in 2022?

Give Claude Code a real browser

nkko — Thu, 02 Apr 2026 19:42:47 +0000

Article URL: https://steel.dev/blog/give-claude-code-a-real-browser

Comments URL: https://news.ycombinator.com/item?id=47619225

Points: 2

# Comments: 0

Startup Is Probably Dead on Arrival

nkko — Wed, 25 Mar 2026 10:32:08 +0000

Article URL: https://steveblank.com/2026/03/17/your-startup-is-probably-dead-on-arrival/

Comments URL: https://news.ycombinator.com/item?id=47515528

Points: 3

# Comments: 0

New comment by nkko in "GitHub explains six February outages and fixes"

nkko — Sat, 14 Mar 2026 13:51:50 +0000

This wqw an interesting postmortem for me. As a regular user, February definitely felt a bit shakier than usual.

I ran into the unicorn error page a few times and had intermittent issues with PR pages and bunch of weirdness. Nothing long-lived, but enough transient failures that my overall feel was it is degrading in general.

GitHub explains six February outages and fixes

nkko — Sat, 14 Mar 2026 13:44:39 +0000

Article URL: https://github.blog/news-insights/company-news/github-availability-report-february-2026/

Comments URL: https://news.ycombinator.com/item?id=47376613

Points: 3

# Comments: 2

New comment by nkko in "Wired headphone sales are exploding"

nkko — Fri, 13 Mar 2026 22:50:06 +0000

was listening to music while coming with groceries and simultaneously juggling stuff to open the doors and change the track with Siri (the only use for Siri I have)

The Post-Copyright Era of Software

nkko — Mon, 09 Mar 2026 21:24:25 +0000

Article URL: https://www.nibzard.com/post-copyright-era-software

Comments URL: https://news.ycombinator.com/item?id=47315757

Points: 5

# Comments: 0

BullshitBench v2: LLMs answering nonsense questions

nkko — Thu, 05 Mar 2026 12:23:27 +0000

Article URL: https://github.com/petergpt/bullshit-benchmark

Comments URL: https://news.ycombinator.com/item?id=47260807

Points: 2

# Comments: 0

Turning web runs into scripts with Codex

nkko — Wed, 04 Mar 2026 18:02:22 +0000

Article URL: https://www.nibzard.com/cashout

Comments URL: https://news.ycombinator.com/item?id=47251333

Points: 1

# Comments: 0

New comment by nkko in "Show HN: Letting Claude automate fleets of browser sandboxes"

nkko — Tue, 03 Mar 2026 18:16:07 +0000

FWIW I work at Steel (not the OP). While we’ve been iterating on the “right shape” for agent tooling, I’ve been building a benchmark harness to measure how different surfaces affect real web task completion: raw API context, CLI-only, opinionated “skills” (structured outputs + artifact capture), and combinations.

If you’ve run agents on the open web, I’d love suggestions for nasty-but-representative workflows to include in the benchmark.

Go vs. Python for AI Infrastructure: Benchmarks 2026

nkko — Tue, 03 Mar 2026 11:32:02 +0000

Article URL: https://dasroot.net/posts/2026/02/go-vs-python-ai-infrastructure-throughput-benchmarks-2026/

Comments URL: https://news.ycombinator.com/item?id=47230946

Points: 1

# Comments: 0

New comment by nkko in "Parallel coding agents with tmux and Markdown specs"

nkko — Tue, 03 Mar 2026 05:57:41 +0000

This rings true, as I’ve noticed that with every new model update, I’m leaving behind full workflows I’ve built. The article is really great, and I do admire the system, even if it is overengineered in places, but it already reads like last quarter’s workflow. Now letting Codex 5.3 xhigh chug for 30 minutes on my super long dictated prompt seems to do the trick. And I’m hearing 5.4 is meaningfully better model. Also for fully autonomous scaffolding of new projects towards the first prototype I have my own version of a very simple Ralph loop that gets feed gpt-pro super spec file.