frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Open in hackernews

What 1k Harness Experiments Taught Me About Self-Improving Agents

https://www.henrypan.com/blog/2026-05-25-self-improvement-harness/
2•megadragon9•47m ago

Comments

megadragon9•47m ago
I recently wanted to see whether an AI agent could self-improve a harness to solve terminal bench tasks. It’s possible for an AI agent to propose a meaningful one-time change to the harness, but after experimenting with this for a couple of weeks, I think the continuous self-improvement is mostly an experiment-systems problem. The system needs a way to decide what kind of improvements can safely compound.

Turns out there's a lot of parallels to coding-agent customization (e.g. SKILLS.md etc..) too.

I wrote my experience of building such system here, including the successful and failure attempts during the process, and how I approached the self-improvement loop. It's not intended as a benchmark claim but more of a systems/research writeup.

https://www.henrypan.com/blog/2026-05-25-self-improvement-ha...

The rise and fall of the only female Yakuza

https://www.theguardian.com/news/2026/may/21/the-devils-child-the-rise-and-fall-of-the-only-femal...
1•NaOH•35s ago•0 comments

Why frontier biology labs need Lisp-like infrastructure

https://www.countifybio.com/
1•mfisc_019•1m ago•0 comments

Some of Texas's oldest barbecue joints close as meat prices skyrocket

https://www.washingtonpost.com/nation/2026/05/25/some-texass-oldest-barbecue-joints-close-meat-pr...
1•paulpauper•1m ago•0 comments

Steam Deck OLED is back in stock, with a price increase for both models

https://store.steampowered.com/news/group/45479024/view/672869045073085538
2•no_news_is•2m ago•0 comments

Agents Thinking Fast and Slow: A Talker-Reasoner Architecture

https://arxiv.org/abs/2410.08328
1•jalcazar•3m ago•0 comments

The AI tech job slaughter gets real

https://www.computerworld.com/article/4175956/the-ai-tech-job-slaughter-gets-real.html
1•CrankyBear•3m ago•0 comments

Remarks on the Disproof of the Unit Distance Conjecture [pdf]

https://cdn.openai.com/pdf/74c24085-19b0-4534-9c90-465b8e29ad73/unit-distance-remarks.pdf
1•digital55•5m ago•0 comments

Datasette: An open source multi-tool for exploring and publishing data

https://datasette.io/
1•Olshansky•6m ago•0 comments

AgentSafeLabs – Launched Open-source Security framework for AI agents

https://github.com/AgentSafeLabs/safelabs-eval
1•waqarjaved•6m ago•0 comments

QuestDB 9.4.0

https://github.com/questdb/questdb/releases/tag/9.4.0
1•tosh•7m ago•0 comments

RTMH: Pope Leo's Magnifica Humanitas on AI

https://thezvi.substack.com/p/rtmh-pope-leos-magnifica-humanitas
1•paulpauper•7m ago•0 comments

Use AI This Election

https://www.astralcodexten.com/p/use-ai-this-election
2•paulpauper•8m ago•0 comments

Multi-Agent LLM System for Automated Vulnerability Discovery and Reproduction

https://arxiv.org/abs/2605.21779
1•root-parent•10m ago•0 comments

Verilog: Back to the building blocks' building blocks

https://www.cs.cornell.edu/~asampson/blog/buildingblocks.html
1•fanf2•10m ago•0 comments

One repo clone, shared forever

https://falconer.com/notes/persistent-repos-s3-files/
2•aryamanagraw•10m ago•0 comments

Benford's Law

https://en.wikipedia.org/wiki/Benford%27s_law
3•jonbaer•15m ago•0 comments

SimCity 3k in 4k

https://www.thran.uk/writ/hdid/2025/12/simcity-3k-in-4k.html
3•speckx•16m ago•0 comments

To Land a Job in AI, Try Reading Kant

https://www.wired.com/story/to-land-a-job-in-ai-try-reading-kant/
2•CharlesW•16m ago•0 comments

Repoprompt is going Open Source

https://repoprompt.com/blog/repo-prompt-next-chapter
1•mirzap•17m ago•0 comments

One Million Beings

https://sub.davidoreilly.com/p/one-million-beings
2•m3at•17m ago•1 comments

Show HN: Citadeld – replay any CI failure locally from a single file

1•hknzerodark1•19m ago•0 comments

CSCI 1377: Tools For Thought (Spring 2026)

https://cel.cs.brown.edu/csci-1377-s26/
2•wcrichton•20m ago•0 comments

Instead of LLMs we need self-updating SLMs

https://crib.social/notice/B6jGNsfie10RYn20lU
1•gslepak•20m ago•0 comments

Show HN: cuSBF – Faster GPU Bloom Filter for Sequence Data

https://github.com/tdortman/cuSBF
2•tdortman•21m ago•0 comments

Objective metrics that change the most as we age

https://www.empirical.health/blog/biomarkers-that-change-with-age/
6•brandonb•21m ago•0 comments

90% cheaper repo inference with GPT-5.4 nano

https://charlielabs.ai/blog/90-percent-cheaper-repo-inference-with-gpt-54-nano/
1•mrbbk•22m ago•0 comments

Vibe-coding a side project while watching TV

https://marindedic.com/swipertab/
5•Realman78•24m ago•1 comments

Attractive faces draw our gaze but fail to hijack our peripheral attention

https://www.psypost.org/attractive-faces-draw-our-gaze-but-fail-to-hijack-our-hidden-attention/
3•Vaslo•26m ago•0 comments

DeepSeek's 10T USD grand strategy

https://twitter.com/i/status/2057909493250539891
5•ssivark•26m ago•0 comments

Context Window Packing – Agent Patterns Catalog

https://www.agentpatternscatalog.org/patterns/context-window-packing/
1•ankitg12•26m ago•0 comments