AdilShamim8/AI-Engineering

A curriculum and field guide for building, evaluating and operating AI systems.
> Evidence review: October 9, 2026. The latest job-data snapshot inspected is September 23, 2026. Review dates and source publication/collection dates are separate. Some long-form legacy claims still need source-level verification; see the audit.
> Section review: October 17, 2026. The AI Evaluation and AI Security & Safety sections were added with all references verified on that date.
## Start here
From foundations to advanced engineering — competency stages, real-data projects, measurable exit criteria and expert review questions.
Evidence and real datasets — a reproduced job-posting analysis, dataset limitations, dated benchmark references and reproduction commands.
Evidence policy — distinguish measurements, source-reported claims, design targets and teaching illustrations.
Repository audit — what was checked, corrected and left unresolved.
The reproduced source contains 8,051 stored job observations across nine scrape folders, including 1,086 records in the September 23, 2026 folder. These are counts from a selected job-board dataset, not worldwide demand estimates. Structured skills and role labels are LLM-derived annotations; the analysis and input fingerprint make the counting reproducible without certifying every annotation.
Advanced practice requires a baseline, versioned data, held-out evaluation, failure analysis, authorization checks, measured operating costs and a recovery plan. The curriculum retains its introductory explanations; the advanced path connects them to reviewable engineering artifacts.
## Roadmaps (Full + Accelerated 100-Day)
Two complete roadmaps for becoming an AI engineer, plus a full interview-preparation track — structure adapted verbatim from the CampusX AI Engineering Roadmap 2026, with every learning resource replaced by free worldwide alternatives:
Roadmap hub — which one to follow — comparison, and how important each topic is for interviews vs the real-world job
Full Roadmap — 16 levels · 7 phases · 286 sessions · 14 projects; 100% coverage from software fundamentals to interview prep
Accelerated Roadmap — 100 Days of AI Engineering — 6 modules · 100 days; the core ~80% you need to land your first AI engineering job
The 100 GenAI / AI Engineer Interview Questions — copied exactly, with strong-answer outlines and a practice loop
12 portfolio projects — full specifications with "done when" criteria and a target-role selection table
## Technical Curriculum (101–106)
101 — LLM Literacy — first-principles guide to transformer mechanics, inference sampling, prompt & context engineering
102 — AI Application Development — orchestration, vector databases, complete RAG curriculum, memory systems, and fine-tuning
103 — Agentic AI — autonomous planning loops, Model Context Protocol (MCP), agentic frameworks, and multi-agent orchestration
104 — LLMOps — deployment architectures, observability, safety guardrails, governance, and enterprise scaling
105 — AI Evaluation — evaluation fundamentals, LLM-as-a-Judge, agent benchmarks and the integrity crisis, RAG evaluation practice, eval engineering in CI/CD, dataset engineering, and online experimentation — plus a complete 19-lecture LLM Evaluations course with full English notes and transcripts, a glossary, a self-assessment quiz, and a cheat sheet