alvinunreal

alvinunreal / awesome-autoresearch

Public

A curated list of autonomous improvement loops, research agents, and autoresearch-style systems inspired by Karpathy's autoresearch.

agentic-systems ai-agents ai-research ai-tools autonomous-agents
245
14
100% credibility
Found Mar 24, 2026 at 245 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
AI Summary

A curated collection of links to projects, papers, and resources on autonomous AI improvement loops and research agents inspired by karpathy/autoresearch.

How It Works

1
🔍 Discover the List

You find Awesome Autoresearch while searching for smart AI tools that improve themselves.

2
📖 Browse Categories

You scroll through fun sections like general tools, research helpers, and special adaptations.

3
💡 Spot Your Match

A project catches your eye, like one for everyday experiments or scientific discovery.

4
Choose Your Path
🛠️
Try General Tools

Jump into versatile self-improving loops for coding or testing ideas.

🔬
Explore Research

Dive into full research agents that plan experiments and write reports.

5
🚀 Visit and Try

Click to a promising project, follow simple guides, and watch the AI learn and get better.

6
⭐ Share or Contribute

Star favorites, read stories from others, or add your own finds to the list.

🎉 Unlock AI Magic

You now have a treasure trove of ideas to make AI agents smarter on their own.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 245 to 245 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is awesome-autoresearch?

This GitHub repo delivers a curated list—your high-signal intel on autonomous improvement loops, research agents, and autoresearch-style systems, all inspired by Karpathy's original autoresearch. It organizes forks, ports, benchmarks, and use cases into clean categories like general-purpose descendants, domain adaptations, and AI scientist tools, saving you hours of scattered GitHub hunting. Built as a Markdown README, it's a static, PR-welcome index under CC0 license, perfect for bookmarking curated programming GitHub gems in agents and self-improvement.

Why is it gaining traction?

Unlike scattered forks or raw paper lists, this awesome curated list stands out with structured sections on platform ports (macOS, WebGPU, Windows RTX), evals like MLAgentBench, and real-world writeups from Shopify to Vesuvius Challenge—making autoresearch accessible beyond H100 clusters. Developers grab it for the "keep-or-revert" pattern generalizations to trading, kernels, or sudoku solvers, plus hooks like distributed GPU swarms and self-evolving agents. At 245 stars, it's pulling devs chasing autonomous ML experiments without the noise.

Who should use this?

AI researchers prototyping end-to-end scientific pipelines, from lit reviews to paper drafts. ML engineers forking autoresearch for consumer hardware or domain tweaks like biomechanics or genealogy. Devs building agent swarms or self-improving coders who need a curated list synonym for quick onboarding to GEPA, ClawTeam, or HGM benchmarks.

Verdict

Solid starting point for autoresearch explorers—star it as curated intel GitHub-style, despite the 1.0% credibility score and modest 245 stars signaling early maturity with no tests or ongoing commits. Pair with originals for production; it's a time-saver, not a runtime.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.