The Agent Harness course

A hand-drawn course following one real agent from a customer's message to the real code.

Active · join meAI Engineering

6 of 9 tasks done

1 · Purpose

This course follows one real agent, the motor-insurance claims assistant on Azure, from the moment a customer types to the moment the reply goes back. It comes in two parts: a hand-drawn tutorial where every box opens the next level down until you reach the code, and a deep-dive Series with each of 22 concerns on its own page. It is for engineers and architects who can call a model but have not yet built everything around it. Real numbers, traces and code come from the live agent, and anything not yet built is labelled so.

For: Software engineers moving into AI, architects, technical leads, study groups.

2 · Business benefits

Faster onboardingNew team members learn a production agent's design in hours instead of reading code for weeks.
Fewer design mistakesEach concern shows what goes wrong without it, with real failures from a live agent.
Cloud choices made clearlyEach concern compares the Azure option with open-source options.
Grounded in real codeCode on each page is cut from the public repositories at build time, so it stays current.

3 · Status and how to join

Active · join me. Under way: pick a task below, or tell me what you'd like to learn and I'll suggest one. Lead: Basant Choudhary.

Good first tasks this week:

  • Read the tutorial start to end and list unclear pages
  • Draft an exercise for one lesson
  • Check the code on three pages still matches the repository

The code is not public yet; joiners get access.

4 · Requirements

What it must do

How well it must do it

5 · Architecture, very high level

Content sourcesThe live agent's repositories,plans and traces.Page registryEach page's id, title, parentand summary, which driveDrawing kitHand-drawn boxes and arrows,each box clickable.Code cutterPulls real code from therepositories by anchor at buildBuilderWrites every page and checksevery link.PublisherRedacts private details andcopies the site to the publicsnippetssite
Components and what each is responsible for, with no tool names: the tools come in the next section.

Also: Content sources → Code cutter: real code · Page registry → Builder: page tree · Drawing kit → Builder: drawings

6 · Tech stack, per platform

ComponentOpen stackAzureAWS
Content sources—the claims agent on Container Apps, API Management, Entra ID and Application Insights ✓not covered
Drawing kitinline SVG ✓——
BuilderPython static site generator ✓——
PublisherGitHub Pages ✓——

Bold with ✓ = built and running today. Everything else is the planned or possible binding: the components stay the same, only the tools change.

7 · What needs to be done

The Series 2 of 3 done

TaskStatusSizeSkill
22 concerns, each on its own pagedoneLAI architecture
Live flow pages from real Azure turnsdoneMObservability
Study the flows not yet covered, such as withdrawing a claimtodoMAI architecture

The tutorial 3 of 3 done

TaskStatusSizeSkill
Hand-drawn pages from architecture down to codedoneLTechnical illustration
Real code, call stacks and class diagrams on design pagesdoneMPython
Tech comparisons for every concerndoneMCloud and open source

Published 1 of 3 done

TaskStatusSizeSkill
Live at /learn/agent-harness/ with a redacted SeriesdoneSWeb publishing
One page per lesson with a short exercisetodoMTeaching
Add a data engineering course in the same styletodoLData engineering

Size: S = an evening or two, M = about a week part-time, L = a few weeks.

8 · Roles

RoleHow manyHours / weekWhat they do
Course reviewer21-2 hours/weekReads pages as a newcomer and reports what was unclear.
Content author13-5 hours/weekWrites and draws new pages from the live agent.
Exercise designer12-3 hours/weekWrites hands-on exercises for each lesson.

9 · What you'll learn

Level-up questions

10 · Links