AI Engineering

AI Engineering Bootcamp

Build production LLM apps with agents, RAG, evals, and deployment workflows.

Level

Intermediate

Duration

64 hours

Lessons

54 lessons

What you will build

Deploy a customer-support AI assistant with tracked evaluations.

Ship a retrieval-augmented assistant with source citations

Design agent workflows with tool calling and guardrails

Evaluate model quality, latency, and refusal behavior

Each module includes practical lessons, time estimates, and project checkpoints.

64 hours total

Understand how modern language models behave before building on top of them.

12h 20m

Prompt architecture
Free preview
Design system, developer, and user prompts with predictable outputs.
1h 40m
Embeddings and semantic search
Generate embeddings, compare vector distance, and reason about recall.
2h 10m
Token budgets and context windows
Control cost, latency, and truncation in long-context workflows.
1h 55m
Model selection tradeoffs
Choose between fast, reasoning, and multimodal models for product tasks.
2h 05m

Build retrieval pipelines that cite sources and handle messy knowledge bases.

18h 10m

Document ingestion
Parse PDFs, pages, and notes into clean chunks with metadata.
2h 20m
Chunking strategy
Compare fixed, semantic, and hierarchical chunking approaches.
2h 05m
Vector search and reranking
Improve relevance with hybrid retrieval and reranking stages.
2h 45m
Citation quality
Return grounded answers with source snippets and confidence states.
2h 15m

Add tool calling, safety checks, and measurable quality gates.

16h 35m

Tool calling
Connect APIs, functions, and structured outputs to agent workflows.
2h 35m
Guardrails and refusal tests
Handle sensitive requests, hallucinations, and unsafe tool usage.
2h 25m
Regression testing
Create test suites for accuracy, latency, and answer consistency.
2h 50m
Deployment checklist
Ship logging, monitoring, fallback behavior, and cost controls.
2h 10m