Skip to content

Day 2 — Patterns & hands-on

Goals for the day

  • Move from single-agent to multi-agent patterns and know when to use which
  • Wire agents to real systems via MCP
  • Make agents stateful with memory; make changes safe with evals

Agenda

TimeSessionFormat
09:30 – 09:45Day 1 recap, parking-lot reviewDiscussion
09:45 – 11:00Multi-agent patterns: supervisor, swarm, pipeline, parallelLecture + demo
11:00 – 11:15Break
11:15 – 12:30Lab 2 — build a supervisor agent that delegates to specialistsHands-on
12:30 – 13:30Lunch
13:30 – 14:30Memory: short-term, long-term, retrieval, compactionLecture + demo
14:30 – 15:30MCP deep-dive — write a custom MCP server, connect itLecture + hands-on
15:30 – 15:45Break
15:45 – 16:45Evals — why they matter more than tests for agentsLecture + lab
16:45 – 17:00Day 2 retroRound-the-room

What participants will leave with

  • A multi-agent system (supervisor + 2 specialists) running locally
  • A working custom MCP server they wrote
  • A first eval suite for one of their agents

Materials

  • Lab 2 starter repo: link TBD
  • MCP server template: link TBD
  • Eval scaffolding: link TBD

Attendees

  • NBS: TBD
  • MetaWeave: TBD

Decisions

  • Captured live.

Action items

  • Owner — action — due