Learning track · 10 weeks, part-time
AI Engineering
Learn the end-to-end lifecycle of building production-grade AI applications. Cover prompts, embeddings, retrieval, RAG evaluation, tools, and basic MLOps.
RAG, vector databases, LLM apps, agents
The difference between a demo and a product is the part after the first prompt: retrieval quality, evaluation, cost, and what happens when a tool call fails.
What you'll cover
Prompting, embeddings, and retrieval
RAG evaluation and improving answers with evidence
Tools, agents, and tool-call failures
Security, cost control, and observability basics
What you ship
Three AI apps: chat, RAG, and tools/agents
Provider and vector store comparisons, with the tradeoffs written down
A project you can demo and defend in an interview
Who it's for
Developers moving from tutorials to AI features, web developers adding LLM capabilities to products they already ship, and data people who want to own the retrieval layer.
FAQ
AI Engineering questions
The questions people ask before starting this track, answered the same way we would in the Discord.
- Do I need machine learning experience?
- No formal ML background is required. The track is applied AI engineering: prompts, embeddings, retrieval, RAG evaluation, tools, and basic MLOps.
- What do I ship?
- Three AI apps: a chat app, a RAG app, and a tool-using agent, plus comparisons of model providers and vector stores.
- How is the 10 weeks structured?
- Part-time, project-first. Alongside the apps you cover security, cost control, and observability basics, so the features hold up outside a demo.
- Which track pairs best with this one?
- Web Development, if you want to ship the full product around the AI feature. Data and Analytics, if you want to work on evaluation and the data layer behind retrieval.