Automate machine-learning research loops by modifying train.py and running fixed 5-minute experiments.
npx skills add https://github.com/jonmumm/skills --skill autoresearch-jonmumm
Please help me install this Agent Skill. Skill: autoresearch Source: https://github.com/jonmumm/skills/tree/main/autoresearch Command: npx skills add https://github.com/jonmumm/skills --skill autoresearch-jonmumm
Autonomous machine-learning research is time-consuming and error-prone when performed manually. This Skill coordinates an AI agent to iterate on a tiny language model by editing train.py via a human-defined program.md, executing short experiments, and aggregating progress overnight.
Open program.md, define the research goals, and start the autonomous loop by running Claude Code to let the agent modify train.py and run fixed 5-minute experiments.
High-intent search queries and answers about installing and using this skill.
You can automate overnight ML research loops by defining research goals in a human-authored program.md file, which instructs an AI agent to iteratively modify train.py and run fixed -minute experiments with automatic commit and revert decisions for reproducible progress.
An autonomous ML research loop uses an AI agent to read human-defined instructions in program.md, make targeted code changes to train.py, execute fixed 5-minute experiments, and log results. Successful changes are committed while failures are reverted, ensuring safe continuous improvement.
Yes, GPU or ML hardware prerequisites are required to execute the short 5-minute experiments and run the autonomous training loops defined by your program.md instructions within the fixed time budget.
You guide the AI agent by writing a detailed research outline in program.md. The agent autonomously executes within your defined 5-minute experiment budgets, modifying train.py and preserving safety checks and audit trails for continuous overnight improvement.
Failed machine-learning training runs are automatically reverted. The system logs each experiment, committing successful changes to train.py and reverting failures, which maintains safety checks and enables continuous reproducible improvement without manual intervention.
akillnessAutomate ML experimentation loops by editing train.py and logging results.
HKUDSSend keystrokes and capture pane output from local tmux sessions.
HarriesAutomate ML/AI research-paper writing with citation verification and LaTeX integration.
doanchienthangdevImplement machine learning training pipelines with PyTorch, Optuna, and MLflow.
