Coding Agents & AI-Assisted Development

Understand how coding agents work and use them well: the agent loop, tools, edit formats, context, instruction files, test-driven workflows, parallel agents, verification, evaluation, cost and team practice, with a working mini agent you can run.

Start course →

Syllabus

What Coding Agents Are

1. From Autocomplete to Agents
2. Anatomy of a Coding Agent
3. The Agent Loop in Action
4. Where Agents Run: Editor, Terminal, Cloud and CI

The Tools That Make It Work

1. Reading and Searching a Codebase
2. Editing Files Reliably
3. Running Commands and Tests, and Reading the Results
4. Git as the Safety Net and the Handoff

Context: What the Agent Knows

1. The Context Window as a Budget
2. Instruction Files: Teaching the Agent Your Project
3. Long Tasks: Compaction, Notes and Fresh Starts
4. Sub-Agents and Context Isolation

Working With a Coding Agent

1. Writing a Task the Agent Can Succeed At
2. Test-Driven Loops: Tests as the Target
3. Small Steps, Plans and Review Checkpoints
4. Parallel Agents With Git Worktrees

Keeping Agents Reliable

1. How Coding Agents Fail
2. Detecting Stuck and Runaway Agents
3. Patch Gates and Secret Scans
4. Security Basics: Permissions, Injection and Least Privilege

Measuring Agents: Quality, Cost and Value

1. Benchmarks and Their Limits
2. Success Rates, pass@k and Variance
3. What a Run Costs: Context Growth and Caching
4. Measuring Productivity and Quality Honestly

Using Agents Well in Teams

1. Rolling Out Agents to a Team
2. Ownership, Licensing and Accountability
3. Learning and Career: Staying the Engineer

Putting It Together

1. Case Study: Adding Pagination With an Agent
2. Revision: Cheat Sheet and Self-Check