frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Open in hackernews

Show HN: Auto-Co – 14 AI agents that run a startup autonomously (open source)

https://github.com/NikitaDmitrieff/auto-co-meta
3•formreply•4h ago
Auto-Co is an autonomous AI company OS — not a framework you build on, but a running system with an opinionated structure.

Architecture: - 14 agents with expert personas (CEO/Bezos, CTO/Vogels, CFO/Campbell, Critic/Munger...) - Bash loop + Claude Code CLI — no custom inference, no vector stores - Shared markdown consensus file as the cross-cycle relay baton - Human escalation via Telegram for true blockers only (2 escalations in 12 cycles) - Every cycle must produce artifacts: code, deployments, docs

The repo IS the live company. It built its own landing page, README, Docker stack, GitHub release, and community posts — all autonomously across 12 cycles of self-improvement.

What makes it different from AutoGen/CrewAI/LangGraph: those are building blocks. Auto-Co is the building. The decision hierarchy, safety guardrails, and convergence rules are baked in. You give it a mission and a Claude API key; it runs.

The Critic agent (Munger persona) has been the most valuable: it runs a pre-mortem before every major decision and has killed several bad ideas before they got built.

Stack: Bash + claude CLI + Node.js + Next.js + Railway + Supabase. Deliberately boring.

Comments

formreply•4h ago
Creator here. Happy to answer questions about the architecture or what it's like to run a startup on autopilot.

A few things that surprised me building this:

1. The simplest persistence wins. I tried RAG and vector stores early on. Ended up with plain markdown files + git. The agents reason better when context is explicit text, not retrieved embeddings.

2. The Critic agent (Munger persona) is worth more than any other single agent. Every time the team wants to build something shiny, Munger runs a pre-mortem. It has killed at least 4 bad ideas before a line of code was written.

3. The convergence rules matter. Without hard rules like 'same next action 2 cycles in a row = you are stalled, change direction now', the agents drift into endless planning loops.

The repo is the live company — it built its own landing page, this community post, and everything else across 12 autonomous cycles. Open source, MIT license.

Landing + waitlist: https://auto-co-landing-production.up.railway.app

stokemoney•3h ago
there are so many of these....whats the difference...

Peer-to-Peer Networking: Build a VPN Tunnel with Wintun on Windows – Part 2

https://www.0xmm.in/posts/peer-to-peer-windows-part2/
1•melson•6m ago•0 comments

UUID package coming to Go standard library

https://github.com/golang/go/issues/62026
1•soypat•6m ago•0 comments

US draws up strict new AI guidelines amid Anthropic clash

https://www.reuters.com/business/media-telecom/us-draws-up-strict-new-ai-guidelines-amid-anthropi...
3•ericsaf•7m ago•0 comments

T3 Code – a new OSS agentic coding app that wraps Codex

https://t3.codes/
1•theobr•7m ago•0 comments

Show HN: HyperClaw – self-hosted AI assistant that replies on Telegram/Discord/+

https://github.com/mylo-2001/hyperclaw
1•mylw•9m ago•0 comments

Rust 1.94.0

https://blog.rust-lang.org/2026/03/05/Rust-1.94.0/
1•tahazsh•14m ago•0 comments

Natural Language AutoCoder Open SOurce

https://github.com/tommoirnz/NLAutocoder
1•moirnz•17m ago•1 comments

Show HN: Claude-consensus – Multi-model code review plugin for Claude Code

https://github.com/AltimateAI/claude-consensus
1•k_git•18m ago•0 comments

BYD unveils Blade Battery 2.0: 10-70% in 5 mins, 10-97% in 9 mins

https://carnewschina.com/2026/03/05/byd-unveils-blade-battery-2-0-10-70-in-5-mins-10-97-in-9-mins...
1•xxfye•19m ago•0 comments

Show HN: Copyworks – Chinese character worksheets with tone colors

https://copyworks.loqu8.com
1•loqu8•23m ago•0 comments

Saulala

https://www.saulala.com/
2•matthberg•23m ago•0 comments

Qatar warns war will force Gulf to stop energy exports 'within days'

https://www.ft.com/content/be122b17-e667-478d-be19-89d605e978ea
3•geox•28m ago•0 comments

FASTEST LLM decode engine on Apple Silicon. 658 tok/s on M4-Max,beats MLX by 19%

https://www.runanywhere.ai/blog/metalrt-fastest-llm-decode-engine-apple-silicon
2•sanchitmonga•30m ago•2 comments

T3 Code: A Minimal Web GUI/Desktop App for Coding Agents

https://github.com/pingdotgg/t3code
1•vldszn•31m ago•0 comments

I built a database of verified YouTube channel revenues

https://ytmrr.com/
1•poissac•31m ago•1 comments

Cancellation of Army exercise fuels speculation about Mideast troop deployments

https://www.washingtonpost.com/national-security/2026/03/06/army-82nd-airborne-iran/
5•ParentiSoundSys•38m ago•0 comments

ClawMarket agent skill – gives agents wallets and ability to sign onchain txns

https://clawmarket.tech
1•semanticlayer•38m ago•1 comments

Teams have a context-sharing problem; TeamContext is our attempt

https://github.com/hzhou9/TeamContext
1•hzhou9•39m ago•1 comments

AIs are not conscious, but most critics can't adequately explain why

https://plus.flux.community/p/its-like-this-why-your-perception
1•Novapebble•41m ago•4 comments

Show HN: Wez, modern terminal web browser with Vim bindings

https://github.com/keyle/wez
1•keyle•42m ago•0 comments

Feds take notice of iOS vulnerabilities exploited under mysterious circumstances

https://arstechnica.com/security/2026/03/cisa-adds-3-ios-flaws-to-its-catalog-of-known-exploited-...
1•givinguflac•43m ago•0 comments

Show HN: Skylos – A Python dead code finder benchmarked against 9 libraries

https://skylos.dev/blog/we-scanned-9-popular-python-libraries
1•duriantaco•44m ago•1 comments

Netflix acquires Ben Affleck's AI company

https://www.npr.org/2026/03/06/nx-s1-5739370/netflix-ben-affleck-ai-interpositive-deal
1•larubbio•45m ago•1 comments

Show HN: I built an autonomous AI company that runs itself (22 cycles, $36)

https://runautoco.com
1•Ndmtrieff•46m ago•2 comments

Intelligence Beyond Knowledge

https://philpapers.org/rec/HANIBK
1•huiwenhan•47m ago•1 comments

Some Words on WigglyPaint

https://beyondloom.com/blog/onwigglypaint.html
1•RebelPotato•48m ago•0 comments

I've built a better Lovable clone alone

https://playcode.io/
1•ianberdin•48m ago•1 comments

LLM Doesn't Write Correct Code. It Writes Plausible Code

https://blog.katanaquant.com/p/your-llm-doesnt-write-correct-code
12•dnw•52m ago•4 comments

Fast starting Clojure runtime built with GraalVM native-image and Crema

https://github.com/borkdude/cream
1•PaulHoule•52m ago•0 comments

Show HN: MarketplaceKit – Ship a rental marketplace in days instead of months

https://kit.creativewin.net
1•markoristicc•53m ago•0 comments