Generative AI Masterclass

From "what is a token?" to a shipped product

Most courses teach Generative AI in disconnected toy examples. This one grows a single real product across every section, so your skills compound instead of resetting. You'll build Sage — a production-shaped AI assistant for the fictional startup Brightpath — one layer at a time, ending with a system that retrieves from private docs, uses tools, generates content, is evaluated and secured, and runs in production.

Six phases, eighteen sections

1. Foundations (Sections 1–4) — get oriented, then build an accurate mental model of AI, ML, generative models, and LLMs: tokens, training, reasoning models, and honest limits.
2. Prompting & APIs (Sections 5–6) — make models behave reliably, then drive them from code and ship Sage Chat.
3. Knowledge & Retrieval (Sections 7–9) — embeddings, vector search, RAG, and frameworks — ground Sage in private knowledge and make it composable.
4. Agents & Customization (Sections 10–12) — agents with tools and MCP, fine-tuning and open models, and multimodal features.
5. Production & Scale (Sections 13–15) — evaluate, secure, deploy, monitor, and design enterprise-grade systems.
6. Portfolio & Career (Sections 16–18) — standalone builds, the integrated capstone, and the path to a GenAI career.

Build, don't just watch

Each section ends with a hands-on checkpoint that adds a concrete capability to Sage. Along the way you ship seven standalone projects plus a capstone, and every section closes with a quiz so you can check your understanding before moving on. A final assessment ties the whole journey together.

Who this is for

Developers and technical builders who want to go from curiosity to competence in Generative AI. You should be comfortable reading basic Python; everything else — prompting, RAG, agents, fine-tuning, multimodal, evaluation, deployment — is taught from the ground up.