ai-ml-engineer

Build, evaluate, and deploy machine-learning and generative-AI systems from data preparation through inference.

2|Updated Aug 21, 2026

Search Tags:#mlops #llm #rag #machine learning #fine-tuning #model evaluation #inference

Ready to run

npx skills add https://github.com/suhanr/next-gen-claude-skills --skill ai-ml-engineer-suhanr

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: ai-ml-engineer
Source: https://github.com/suhanr/next-gen-claude-skills/tree/main/skills/03-ai-ml-engineer
Command: npx skills add https://github.com/suhanr/next-gen-claude-skills --skill ai-ml-engineer-suhanr

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? It guides the full machine-learning lifecycle so models are built on trustworthy data, evaluated against representative and adversarial cases, and deployed with monitoring instead of shipping untested experiments. ## Core Features & Use Cases - Data-to-Model Lifecycle: Covers dataset strategy, leakage checks, baselines, experiment tracking, and reproducible model versioning. - LLM and Retrieval Engineering: Supports fine-tuning decisions, embeddings, retrieval, reranking, and hybrid search for generative-AI applications. - Evaluation and Serving: Designs evaluation reports, error analysis, inference optimization, and monitoring for drift, latency, cost, and safety. - Use Case: When building a RAG-based support chatbot, use this Skill to compare prompting versus fine-tuning, design an evaluation set, and package inference behind a stable interface with fallbacks. ## Quick Start Ask the AI to act as an ML engineer and design an evaluation and deployment plan for your model or LLM application.

Frequently Asked Questions about ai-ml-engineer

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

FAQPage Schema

How do I decide between fine-tuning, RAG, and prompting for an LLM application?▼

Compare prompting, retrieval, tool use, and fine-tuning based on evidence from a baseline experiment. Start with the simplest approach, evaluate on representative cases, and only add fine-tuning or retrieval complexity when measurements justify it.

How do I evaluate a machine learning model before deployment?▼

Evaluate on representative and adversarial cases rather than average accuracy alone. Establish a trustworthy dataset split with leakage checks, build a baseline first, and produce an evaluation report with error analysis and calibration.

What should I monitor after deploying an ML model to production?▼

Monitor quality drift, data drift, latency, cost, and safety after deployment. Package inference behind a stable interface with timeouts, fallbacks, and observability so regressions are detected early.

When should I not add complexity to an ML system?▼

Avoid adding complexity without a measured requirement. Build a baseline first, prefer the simplest option that satisfies the requirement, and never optimize without measurement or claim production readiness without evidence.

How do I keep ML experiments reproducible?▼

Track experiments, model versions, data versions, and configuration throughout the lifecycle. Use data/version management practices so results can be reproduced and remaining limitations are clearly documented.