skills.rest
htlin222
htlin222COMMUNITY
Explore all skills by this author →
Website𝕏@htlin222GitHub

ml-engineer

Implement machine learning pipelines with FastAPI serving and scikit-learn preprocessing.

77|4|Updated Jun 20, 2022
Search Tags:#fastapi#mlops#machine learning#feature engineering#model deployment#data drift#batch inference
One-click install
npx skills add https://github.com/htlin222/dotfiles --skill ml-engineer-htlin222
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ml-engineer
Source: https://github.com/htlin222/dotfiles/tree/main/claude.symlink/skills/ml-engineer
Command: npx skills add https://github.com/htlin222/dotfiles --skill ml-engineer-htlin222

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires fastapi, joblib, numpy, pandas, scikit-learn, evidently, hashlib, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill streamlines the process of deploying machine learning models into production, handling feature engineering, and monitoring model performance.

Core Features & Use Cases

  • Model Serving: Deploy models as FastAPI endpoints for real-time predictions or implement batch inference for large datasets.
  • Feature Engineering: Construct robust data preprocessing pipelines using scikit-learn for both numerical and categorical features.
  • Model Monitoring: Implement data drift detection and dataset summary reports using Evidently.
  • A/B Testing: Facilitate controlled rollouts and comparisons of different model versions.
  • Use Case: Deploy a trained classification model as a microservice, engineer new features from raw input data, and set up a system to monitor for data drift between training and production data.

Quick Start

Use the ml-engineer skill to deploy a model as a FastAPI endpoint.

Frequently Asked Questions about ml-engineer

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

FAQPage Schema
How do I deploy a scikit-learn model as a FastAPI endpoint for real-time predictions?▼

Deploy a scikit-learn model as a FastAPI endpoint by wrapping the trained pipeline in a web service for real-time predictions. This approach serves machine learning models directly as microservices for immediate inference requests.

What is the best way to detect data drift in production machine learning models?▼

Detect data drift in production machine learning models by generating dataset summary reports using Evidently. This monitors statistical differences between training and production data to maintain model performance over time.

How do I implement concurrent batch inference for large datasets?▼

Implement concurrent batch inference to process large datasets efficiently by applying the trained machine learning pipeline to multiple data records simultaneously. This enables scalable batch predictions for high-volume workloads.

Can I build data preprocessing pipelines for both numerical and categorical features using scikit-learn?▼

Build robust data preprocessing pipelines using scikit-learn to transform both numerical and categorical features. This constructs consistent feature engineering steps for model integration and production deployment workflows.

How does deterministic A/B testing work for comparing different model versions?▼

Deterministic A/B testing for comparing model versions uses hashing to consistently route incoming requests to different model variants. This facilitates controlled rollouts and accurate performance comparisons between deployed machine learning models.

Do I need joblib to serve machine learning models in production?▼

Joblib is required to serialize and deserialize trained scikit-learn pipelines for production deployment. It loads the persisted machine learning models into memory for FastAPI serving or batch inference.

Related Skills

Explore More Software Engineering Skills→
SciMate-AISciMate-AI
1

scikit-learn

Train and evaluate ML models using scikit-learn pipelines.

Official
Advanced
nirukk52nirukk52
1

dev-log-monitoring

Monitor Encore.ts and SvelteKit logs to diagnose drift and verify run events.

Community
Advanced
StarlitnightlyStarlitnightly
1.2k

data-transform

Transform raw tabular data into cleaned, structured datasets using pandas and numpy.

Community
Advanced
luokai25luokai25

mlops-expert

Set up and monitor ML pipelines with MLflow, DVC, FastAPI, and Feast.

Community
Advanced