← Back to catalogue

Discipline herald — Artificial Intelligence
Discipline herald

Level 2 · M.Sc

MSc Applied AI Systems

Artificial Intelligence · Live · v0

Applied AI systems end to end — mathematical foundations, model architectures, training, and production deployment.

Deadline
Next cohort starts 2026-09-07 (AY 2026/27)
Length
42 weeks · 42 phases
Cortex Credits (CC)
126 CC
What are Cortex Credits?
Syllabus
View the phase syllabus
  1. Phase 00 — Foundations
  2. Phase 01 — Hardware & Computing Substrate
  3. Phase 02 — Numerical Representation
  4. Phase 03 — Linear Algebra from First Principles
  5. Phase 04 — Calculus & Optimization for AI
  6. Phase 05 — Probability & Information Theory
  7. Phase 06 — Python for AI Engineering
  8. Phase 07 — Scalar Autograd from Scratch (`minigrad.scalar`)
  9. Phase 08 — Tensor Autograd from Scratch
  10. Phase 09 — MLP, Modules, and Optimizers
  11. Phase 10 — Initialization, Normalization, Residuals
  12. Phase 11 — Tokenization Theory + BPE Implementation
  13. Phase 12 — The Corpus: Designing the Microscopic Dataset
  14. Phase 13 — Embeddings & Representation Spaces
  15. Phase 14 — Pre-Transformer Sequence Models
  16. Phase 15 — Attention from Scratch
  17. Phase 16 — Positional Encodings
  18. Phase 17 — Tiny Transformer Block & Mini-GPT
  19. Phase 18 — Training Loop, Checkpointing, Mixed-Precision Preview
  20. Phase 19 — Training Dynamics & Debugging
  21. Phase 20 — Evaluation Harness
  22. Phase 21 — Inference Internals & Sampling
  23. Phase 22 — KV Cache: From Math to Memory
  24. Phase 23 — GPU Architecture Fundamentals
  25. Phase 24 — CUDA & Triton Hands-On
  26. Phase 25 — PyTorch Internals
  27. Phase 26 — Quantization Deep Dive
  28. Phase 27 — Modern Attention Optimizations
  29. Phase 28 — Fine-Tuning, LoRA, QLoRA
  30. Phase 29 — Retrieval-Augmented Generation (RAG)
  31. Phase 30 — Structured Generation & Constrained Decoding
  32. Phase 31 — Tool Use & the Model Context Protocol (MCP)
  33. Phase 32 — Agents: Planning, Memory, Sandboxing (Capstone Application)
  34. Phase 33 — Inference Serving: From FastAPI to Continuous Batching
  35. Phase 34 — Observability, Cost & Capacity
  36. Phase 35 — Distributed Training & Inference
  37. Phase 36 — Frontier Architectures
  38. Phase 37 — Security & Safety of AI Systems
  39. Phase 38 — MLOps
  40. Phase 39 — Capstone: The Miniature Production System
  41. Phase 40 — Hardening, Postmortem, "What's Next"
  42. Phase 41 — Learner Portal: Delivering the Curriculum to Many
Enrolment prerequisites
  • A verified account and admissions-committee approval.

Professor: Borja Tarraso — Founder & Chief

Sign in to request enrolment