Week 3 of 8 · Build AI

Agents, retrieval and the browser

Agents flow from Week 2's context/memory. Then give agents knowledge (RAG) and capabilities (scraping, browser).

What the week builds toward

WorkBot's investigation agent: an orchestrator delegating to an agentic-retrieval specialist and a browser specialist, with error recovery, structured logging, cost tracking and end-to-end tests.

The 5 days

  1. Day 11

    AI agent fundamentals

    The agent loop and its patterns, then a ReAct research agent from scratch with iteration, cost and time guards.

    You ship: A ReAct research agent from scratch, streaming its reasoning.

  2. Day 12

    Multi-agent systems and frameworks

    Design multi-agent architectures and build a research, writing and editing pipeline on an agent framework.

    You ship: A 4-agent content pipeline with guardrails, handoffs and tracing.

  3. Day 13

    Modern RAG: agentic retrieval

    See where one-shot retrieval fails and build agentic retrieval over a 50-document corpus with a virtual filesystem the agent can explore.

    You ship: Agentic retrieval over a 50-document corpus, benchmarked against top-k RAG.

  4. Day 14

    AI scraping, browser use and desktop use

    Extract structured data from the web with AI, drive a browser from screenshots with a vision model, and understand how computer use works and where its limits are.

    You ship: An AI scraper with pagination and a browser agent that completes a goal.

  5. Day 15

    The research analyst: orchestration, retrieval and the browser

    Combine multi-agent orchestration, agentic retrieval and browser automation into one tested, logged research analyst.

    You ship: The AI research analyst system, with error recovery, logging, cost tracking and tests.