The AI Path

The goal: stop treating AI models as magic boxes and stop learning Rust as disconnected theory. Learn how inference runs, then build useful products where ownership, async, RAG, multimodal evidence, agents, evals, and operations appear exactly when the product needs them.

Part 1 · The mechanics — AI Systems · First Principles

1. Foundations — what an LLM actually costs in latency, memory, and dollars, and why efficiency is the whole game.
2. The transformer, mechanically — attention, prefill vs decode, and why the KV cache exists.
3. Inference efficiency — batching, PagedAttention/vLLM, FlashAttention, speculative decoding — the levers that make it cheap.
4. Application layer + the Claude-API capstone — prompt caching, routing, RAG, and a real before/after cost dashboard.

Part 2 · The design — System Design · First Principles

5. Building With LLMs — how LLM-backed features change system design: RAG, vector DBs, the new latency/cost/non-determinism budget, streaming UX, and guardrails.
6. The data-intensive frontier — vector search (HNSW/IVF), streaming, and the lakehouse that feed AI features.

Part 3 · The build — Rust · Project-First

7. Project 4 · askr, an AI CLI — build a real AI command-line tool on the Claude API: streaming, a tiny RAG (embeddings + vector search), and production hardening. The mechanics and design, made concrete in code.

Part 4 · The complete build — Rust + AI · Project-First

8. Brandforge — turn website evidence into a typed profile, claim audit, image prompt, and visual ad preview.
9. AskDocs — build RAG from files, embeddings, vector search, streaming, citations, and evals.
10. Signal Room + Cutroom — align images, audio, logs, and video into inspectable evidence timelines.
11. Forge Agent — study the Rust Codex CLI and build typed tools, authorization, bounded loops, replay, memory, and release evals.
12. Model Lab + Mosaic — run local models, route and improve them, then ship the durable, secure, observable production platform.

Where it leaves you

You’ll understand an AI run from the artifact backward: what evidence entered, which Rust boundary owned each decision, what the model contributed, how the result was verified, and how the complete service is operated and rolled back.