A free, opinionated 10-month curriculum for becoming a job-ready AI Engineer — built in public by a CS student doing the same journey.
I'm Aryav Chanduka, a third-year B.Tech CSE (AI/ML) student at Manipal University Jaipur. When I started planning how to go from "knows Python" to "AI Engineer who can build and deploy real systems," I couldn't find a single curriculum that covered the full stack — math, classical ML, deep learning, transformers, RAG, fine-tuning, agents, MLOps, system design, and interview prep — in the right order, with the right resources.
So I built one. 40 weeks. 5 phases. Day-by-day tasks with hand-picked resources (mostly Karpathy, fast.ai, Stanford CS229/CS224N, and the original papers). Then I built this tracker so I could check off tasks, see my streak, and stay accountable.
It's free. It's open. If you're also trying to become an AI Engineer, use it.
Phase 1 — Weeks 1–8
Math foundations: linear algebra, calculus, probability, Python ML toolchain, classical ML
Phase 2 — Weeks 9–16
Deep learning with PyTorch: neural nets from scratch, CNNs, NLP, transformers (Karpathy's GPT)
Phase 3 — Weeks 17–24
Applied LLM engineering: prompt engineering, RAG, LoRA fine-tuning, AI agents, production serving
Phase 4 — Weeks 25–32
ML system design and interview prep: DSA, coding patterns, ML breadth, system design cases
Phase 5 — Weeks 33–40
Specialization: computer vision, advanced NLP, internship/MS application prep, portfolio projects
Questions, feedback, or corrections: aryavchanduka18@gmail.com