skills.rest
FMurray
FMurrayCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

mlflow-onboarding

Guides MLflow onboarding by detecting GenAI or traditional ML use cases and integrating tracing or tracking.

Updated Mar 5, 2026
Search Tags:#onboarding#tracing#experiment tracking#mlflow#genai#autologging#llm observability
One-click install
npx skills add https://github.com/FMurray/mlfts --skill mlflow-onboarding-fmurray
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: mlflow-onboarding
Source: https://github.com/FMurray/mlfts/tree/main/.agents/skills/mlflow-onboarding
Command: npx skills add https://github.com/FMurray/mlfts --skill mlflow-onboarding-fmurray

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires mlflow.

What problem does it solve? Setting up MLflow correctly depends on whether you are building a GenAI application or a traditional ML model, and choosing the wrong path wastes time on irrelevant tutorials and integration steps. This Skill determines your use case and walks you through the right quickstart and integration. ## Core Features & Use Cases - Use Case Detection: Scans your codebase for framework imports (openai, langchain, sklearn, torch) and checks the experiment's mlflow.experimentKind tag to classify GenAI vs traditional ML. - Guided Quickstarts: Recommends the relevant MLflow tutorials for tracing, evaluation, prompt versioning, experiment tracking, or hyperparameter tuning. - Project Integration: Adds autologging calls (mlflow.openai.autolog, mlflow.sklearn.autolog, etc.), experiment configuration, and custom tracing or manual logging to your code, then verifies traces or runs appear. - Use Case: You inherit a Python project and ask to add MLflow. The Skill detects LangChain imports, classifies it as GenAI, adds mlflow.langchain.autolog() at the entry point, and verifies traces are captured. ## Quick Start Ask the assistant to help you get started with MLflow in your current project and let it detect your use case automatically.

Frequently Asked Questions about mlflow-onboarding

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I get started with MLflow in my project?▼

First determine whether your project is GenAI (LLM apps, agents) or traditional ML (sklearn, PyTorch). GenAI projects start with tracing via autologging, while ML projects start with experiment tracking. The Skill detects this from your imports and guides the integration.

How do I add MLflow tracing to an LLM application?▼

Add a single autologging call such as mlflow.openai.autolog() or mlflow.langchain.autolog() at your application entry point before any LLM calls. For unsupported providers or custom logic, use the @mlflow.trace decorator on relevant functions.

Does MLflow autologging support my ML framework?▼

MLflow autologging supports scikit-learn, PyTorch, TensorFlow, Keras, XGBoost, and LightGBM, among others. If your framework is unsupported, use mlflow.log_param(), mlflow.log_metric(), and mlflow.log_artifact() for manual logging.

How do I verify MLflow integration is working?▼

For GenAI apps, run the application and search traces with mlflow traces search filtered by experiment ID. For ML projects, run training and check runs with mlflow runs search. Results can also be inspected visually in the MLflow UI.

Can I try MLflow tracing without an LLM API key?▼

Yes. You can create traces with mock data using @mlflow.trace and mlflow.start_span() to simulate LLM calls and retrieval steps. This demonstrates tracing structure without requiring OpenAI or Anthropic credentials.

Related Skills

Explore More Software Engineering Skills→
databricksdatabricks
183

quickstart

Configure Databricks CLI authentication and MLflow tracking in a .env file.

Official
Basic
JingyiWngJingyiWng

databricks-mlflow-evaluation

Evaluate MLflow GenAI agents with built-in and custom metrics.

Community
Advanced
databricks-solutionsdatabricks-solutions
5

prompt-registry-patterns

Implement MLflow Prompt Registry patterns for versioned prompt management with Unity Catalog storage.

Official
Advanced
doanchienthangdevdoanchienthangdev
4

observability

Configure structured logging, Prometheus metrics, OpenTelemetry tracing, and Sentry error tracking.

Community
Advanced