Application Developers

Tick units as you finish them. Progress is saved only in this browser — no account needed.

1. What AI, ML, Deep Learning and LLMs Are

   Stage 0 · 3 h · 3 h total · lesson ready

   One map of the AI family, and how an LLM “predicts the next token”.

   I finished this

2. Prompting & Using AI Well

   Stage 0 · 3 h · 6 h total

   Prompting techniques: role, context, examples, step-by-step thinking and structured output.

   I finished this

3. Python, NumPy & Tooling

   Stage 1 · 12 h · 18 h total

   Read and write Python, vectorise with NumPy, use the command line, Git and Jupyter.

   I finished this

4. PyTorch & Autograd

   Stage 1 · 10 h · 28 h total

   Tensors, autograd, data loading and the training loop.

   I finished this

5. Tokenization & Embeddings

   Stage 4 · 6 h · 34 h total · lesson ready

   BPE tokenisation, per-token pricing, embeddings and semantic similarity.

   I finished this

6. Transformers & Attention

   Stage 4 · 16 h · 50 h total · lesson ready

   Self-attention, multi-head attention, positional encoding, residuals and normalisation — build a GPT from scratch.

   I finished this

7. Instruction Tuning & PEFT

   Stage 5 · 10 h · 60 h total · lesson ready

   SFT, synthetic data, LoRA/QLoRA/DoRA, adapters and model merging.

   I finished this

8. Decoding, Sampling & KV Cache

   Stage 6 · 6 h · 66 h total · lesson ready

   Prefill and decode, temperature/top-p, KV cache and the memory-bandwidth bottleneck.

   I finished this

9. Serving Engines

   Stage 6 · 8 h · 74 h total

   PagedAttention, continuous batching, prefix caching, speculative decoding, prefill/decode disaggregation and gateway routing.

   I finished this

10. Running LLMs Locally

    Stage 6 · 4 h · 78 h total

    llama.cpp, Ollama, MLX, memory estimates and choosing a model.

    I finished this

11. Embeddings, Vector Search & Reranking

    Stage 8 · 8 h · 86 h total

    BM25, sentence embeddings, DPR, ColBERT, HNSW, PQ and hybrid search.

    I finished this

12. RAG Engineering & Evaluation

    Stage 8 · 8 h · 94 h total · lesson ready

    Chunking, query rewriting, HyDE, GraphRAG, citations and evaluation metrics.

    I finished this

13. Agents: Tool Use, MCP, Planning & Memory

    Stage 8 · 12 h · 106 h total · lesson ready

    ReAct, function calling, MCP, reflection, memory, multi-agent systems, computer use and sandboxes.

    I finished this

14. AI Engineering: Evals, Cost & Shipping

    Stage 8 · 8 h · 114 h total

    Your own eval sets, LLM judges, observability, model gateways, cost control and product design.

    I finished this

15. Evaluation & Benchmarks

    Stage 9 · 6 h · 120 h total · lesson ready

    Major benchmarks, arenas, data contamination and the limits of capability evaluation.

    I finished this

16. Security: Jailbreaks, Injection & Privacy

    Stage 9 · 6 h · 126 h total · lesson ready

    Adversarial examples, GCG, prompt injection, red-teaming, watermarking, differential privacy and backdoors.

    I finished this

• 1Everyday Users · No code: in 2–3 weeks, learn to understand and use AI well and spot hallucinations and risks.40h
• 2Algorithm Researchers · From maths to frontier papers: in 6–12 months, read and reproduce papers on LLMs, alignment and reasoning models.257h
• 3Product Managers & Leaders · Understand technical limits, cost structure and risk well enough to choose tools, start projects and set governance.84h
• 4Domain Scientists · With a background in your own field, learn to do research with ML and LLMs — linked with the “AI × Disciplines” section.143h