Skip to content
Projects

AdilShamim8/GenAI-Roadmap-with-Notes-and-Projects

View on GitHub

AdilShamim8/GenAI-Roadmap-with-Notes-and-Projects has 105 GitHub stars and 34 forks. GitHot measured 0 stars in 24 hours. Last checked 2026-11-16 19:47 UTC. Last push 2026-10-20.

How these counts are measured

Measured star growth

in 24h
0
in 7 days
Collecting history
in 15 days
Collecting history
in 1 month
Collecting history

Net star changes over overlapping periods, not daily totals. Missing measurements are left blank. Stars measure attention, not software quality.

About

A comprehensive learning path and practical guide for Generative AI development with hands-on LangChain implementations and detailed notes.

AI

README.md

Original

GenAI Engineering Roadmap

The production-grade, evals-first, career-aware roadmap for GenAI engineers in 2026.
From "I can call the OpenAI API" to "I shipped a multi-agent system that runs in production with SLOs and an evals harness."

License Last verified Changelog CI


What this is

A rigorously-sourced, file-per-topic, evals-first roadmap for becoming a production GenAI engineer. Every factual claim is cited and dated in LAST_VERIFIED.md. Every topic follows a consistent template. Every project ships with code, tests, and an evals harness.

This is not a link dump. This is not a single 600-line README. This is not LangChain-only. This is the roadmap a principal AI engineer would recommend without caveats.

What this is not

  • Not a tutorial. Topics assume you can read Python and understand HTTP.
  • Not exhaustive. We picked one lane — production GenAI engineering — and go deep.
  • Not a model leaderboard. We cover architectures and patterns that survive model releases.

The 12-rung skill ladder

Rung Topic Folder Est. time
0 Foundations 00-foundations/ 1 week
1 Applied LLM Calling 01-llm-calling/ 1 week
2 Context Engineering 02-context-engineering/ 2 weeks
3 Modern RAG 03-rag/ 3 weeks
4 Tools & MCP 04-tools-and-mcp/ 2 weeks
5 Single-Agent Engineering 05-agents/ 2 weeks
6 Multi-Agent Orchestration 06-multi-agent/ 3 weeks
7 Evals & Test-Driven LLM Dev 07-evals/ 2 weeks
8 Production & Economics 08-production/ 2 weeks
9 Fine-tuning & Self-Hosting 09-fine-tuning-and-self-hosting/ 2 weeks
10 Safety, Alignment & Red-Teaming 10-safety/ 1 week
11 Real-Time & Voice 11-realtime-voice/ 1 week

Folders are numbered by rung — 00-foundations/ through 11-realtime-voice/ — so the order is always visible at a glance: work them in numerical order. Supporting material (learning paths, decision guides, projects, curated resources, career guide) lives outside the ladder in docs/, projects/, awesome/, and field-guide/.

Pick your starting point

Your background Recommended path
Backend engineer docs/learning-paths/from-backend-engineer.md
Frontend engineer docs/learning-paths/from-frontend-engineer.md
Data engineer docs/learning-paths/from-data-engineer.md
Data scientist docs/learning-paths/from-data-scientist.md
ML engineer docs/learning-paths/from-ml-engineer.md
Starting from zero docs/learning-paths/from-zero.md

The GenAI Field Guide

Career and industry context — separate from the technical curriculum.

Portfolio projects

15 portfolio-grade builds, each with README.md, src/, tests/, 07-evals/, DEPLOYMENT.md. See projects/.

Decision guides

When you hit a fork in the road, start here:

Curated external resources

awesome/ — papers, books, courses, blogs, tools, communities. One-line reviews, not bare links.

How to use this repo

  1. Read LAST_VERIFIED.md to see when every factual claim was last checked.
  2. Pick a learning path based on your background.
  3. Work the rungs in order. Each rung links to the next.
  4. Build at least 3 projects as you go. Code beats reading.
  5. Run the evals that ship with each project. Internalize the evals-first mindset.

Contributing

See CONTRIBUTING.md. The short version: every factual claim must be sourced and dated, every file must follow the topic template, every project must ship with tests.

License

MIT — see LICENSE.

Acknowledgements

  • alexeygrigorev/ai-engineering-field-guide for the field-guide playbook and the "data, not filler" philosophy.
  • Every practitioner who has published a postmortem, evals harness, or agent architecture in public.

Star history

If this roadmap helps you, star it. Stars keep the project maintained.

✦ Connect With Me

Portfolio LinkedIn Medium Substack

GitHub Kaggle LeetCode X

Building • Learning • Researching • Sharing


⭐ If this repository helped you, consider giving it a star! ⭐