Part of Research and Analysis

Which agents run and log experiments without supervision?

2 examples

Agents

AlphaEvolve
Claude Code

Repositories

karpathy/autoresearch
anthropics/claude-code

Use Cases

Reported by source

Evolve faster algorithms with AlphaEvolve and automated evaluators

Google DeepMind's AlphaEvolve pairs Gemini models with automatic scoring to evolve code, and reports gains in data-centre scheduling, AI kernels and open maths problems.

AlphaEvolve

deepmind.google · 14 May 2025

Reported by source

Run unattended ML training experiments with autoresearch

Red Hat's Roy Belio containerized Karpathy's autoresearch on OpenShift AI with H100 GPUs and let Claude Code run 198 training experiments over 24 hours.

Claude Code

developers.redhat.com · 7 Apr 2026karpathy/autoresearch, anthropics/claude-code

Which tools build a searchable knowledge base from your documents?
Which agents investigate a question and report back?
Which agents research and benchmark code libraries?
Which agents search for faster algorithms or new mathematical constructions?
Which agents help people write software that controls robots or other hardware?

Agents

Repositories

Use Cases

Evolve faster algorithms with AlphaEvolve and automated evaluators

Run unattended ML training experiments with autoresearch

Specific jobs