#140140th of 143 in Coding agents and AI editors
mle-agent
CLI agent that builds ML baselines and runs Kaggle tasks
- Stars
- 1.6k
- License
- MIT
- Last commit
- Jul 2026
- Last release
- Oct 2024
- Language
- Python
Missed the latest GitHub check; these facts may be older.
Overview
MLE-Agent is a Python CLI that acts as a pairing agent for ML engineers. It plans an ML task from vague requirements, writes and runs code locally, debugs errors, and can search Arxiv and Papers with Code. It also has an auto mode for Kaggle competitions, an interactive chat, and a weekly report generator from GitHub or a local git repo.
Who it is for: ML engineers and researchers who want an agent for baselines and Kaggle
Strengths
- Auto Kaggle mode runs data prep to submission without human input
- Searches Arxiv and Papers with Code to inform solutions
- Installs from PyPI with pip or uv; MIT license
- Report command works from GitHub or a local git repo
Weaknesses
- Last tagged release is 2024-10-12; README milestones stop at 0.4.2
- Many roadmap integrations (Hugging Face, MLflow, Wandb, S3) are unchecked
- README names few models: Ollama Llama3, OpenAI, Claude 3.5 Sonnet, Mistral
- Hardware needs and API key setup are not documented in the README
What it needs
- Needs Python, LLM provider or Ollama, Kaggle account (for Kaggle mode)
- Models: OpenAI, Anthropic Claude, Ollama, Mistral
- port 3000
Also in Coding agents and AI editors
See all 143| Rank | Project | Score |
|---|---|---|
| 1 | ArchonYAML workflow engine that runs AI coding agents through fixed dev steps | 84 out of 100 |
| 2 | claw-codeRust CLI coding agent harness using Anthropic and OpenAI-compatible APIs | 83 out of 100 |
| 3 | Open SWEAgent that turns issues into pull requests, reviews PRs and monitors CI | 83 out of 100 |
| #4 | piTerminal coding agent with extensions, skills, and an SDK | 80 out of 100 |
| #5 | oh-my-openagentTerminal coding agent with multi-model orchestration and git-backed memory | 80 out of 100 |