# mlops-mlflow

A skill that gives Claude working instructions for MLflow: logging runs, versioning models in the registry, and reproducing experiments.

- skill · `type=skill` · Claude Code · 64 installs
- Author: Orchestra Research
- License: MIT · policy=link-only · description only; the file stays with the author
- Source: https://github.com/davila7/claude-code-templates/tree/HEAD/cli-tool/components/skills/ai-research/mlops-mlflow
- Updated: 2026-10-24
- HTML: https://affpapa.org/en/nevibecode/skills/mlops-mlflow

## What it does

The skill treats MLflow as an ML lifecycle platform and wires it into the agent's workflow. It ships Python examples: opening a run with `mlflow.start_run()`, recording parameters, metrics and artifacts, saving models from PyTorch, scikit-learn, Keras and HuggingFace Transformers.

Autologging gets its own section. A single `mlflow.autolog()` call, or a framework-specific one like `mlflow.sklearn.autolog()`, captures hyperparameters, metrics and the model, plus training duration. Model registry coverage includes versioning, stage transitions, and registering a model at log time with `registered_model_name`.

The examples assume `mlflow`, `sqlalchemy` and `boto3` are already installed — skip them and the registry and artifact-storage snippets will not run. The local UI starts with `mlflow ui` and serves at `http://localhost:5000`.

## Install

### Claude Code (native platform)

1. **claude-code-templates**
```bash
npx claude-code-templates@latest --skill ai-research/mlops-mlflow
```

2. **Skills CLI (Vercel)**
```bash
npx skills add davila7/claude-code-templates --skill mlops-mlflow -a claude-code
```

3. **Manually: put the files in place**
- File: `~/.claude/skills/mlops-mlflow/SKILL.md` — Copy the whole skill folder; for one project use .claude/skills/mlops-mlflow/

### Codex

**Manually: put the files in place**
- File: `.agents/skills/mlops-mlflow/SKILL.md` — For all projects use ~/.agents/skills/

### Cursor

**Manually: put the files in place**
- File: `.cursor/skills/mlops-mlflow/SKILL.md` — Cursor also reads .agents/skills/ and .claude/skills/

### GitHub Copilot

1. **GitHub CLI**
```bash
gh skill install davila7/claude-code-templates mlops-mlflow
```

2. **Manually: put the files in place**
- File: `.github/skills/mlops-mlflow/SKILL.md`

### Gemini CLI

**Platform CLI**
```bash
gemini skills install https://github.com/davila7/claude-code-templates --path cli-tool/components/skills/ai-research/mlops-mlflow
```

### Windsurf (Devin Desktop)

**Manually: put the files in place**
- File: `.devin/skills/mlops-mlflow/SKILL.md` — The legacy .windsurf/skills/ path still works

### Claude.ai

**Claude.ai**
- Zip the folder and upload it in Settings → Capabilities → Skills

Where a file is given instead of a command, copy the files from the source to that path: https://github.com/davila7/claude-code-templates/tree/HEAD/cli-tool/components/skills/ai-research/mlops-mlflow

## When to use it
- record parameters, metrics and artifacts for every training run
- compare model versions and their scores across runs
- register a model and move it between stages
- reproduce an experiment from a saved configuration
- turn on autologging for scikit-learn or PyTorch instead of logging by hand

## Example prompts
- Wrap the training loop in `mlflow.start_run()` and log learning_rate and batch_size
- Enable `mlflow.sklearn.autolog()` and fit a RandomForestClassifier on the train split
- Register the model as `my-classifier` with `registered_model_name` while logging

## Similar
- [mlops-workflows](https://affpapa.org/en/nevibecode/skills/mlops-workflows.md) — skill · 444 installs
- [mlops-engineer](https://affpapa.org/en/nevibecode/agents/mlops-engineer.md) — subagent · 1009 installs
- [mlflow-onboarding](https://affpapa.org/en/nevibecode/skills/mlflow-onboarding.md) — skill · 1100 installs
- [ml-pipeline-automation](https://affpapa.org/en/nevibecode/skills/ml-pipeline-automation.md) — skill · 389 installs
- [experiment-tracking-swanlab](https://affpapa.org/en/nevibecode/skills/experiment-tracking-swanlab.md) — skill · 896 installs
- [ml-pipeline](https://affpapa.org/en/nevibecode/skills/ml-pipeline.md) — skill · 3900 installs

From the same set: [davila7/claude-code-templates](https://affpapa.org/en/nevibecode/by/davila7/claude-code-templates.md) — 361 cards

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md