ADVANCED · 6 MODULES · 48 LESSONS
Working with Coding Agents
Direct coding agents as a craft: write verifiable specs, set up your repo so the agent can check its own work, and review what it delivers before you accept it.
- For you if
- You already use a coding agent and want to move from improvising to working with a method.
- By the end
- You'll know which tasks to delegate, how to control token and time spend, and how to write your team's agent rules.
- You need
- Experience with a coding agent, Git with pull requests, and a language with automated tests.

Curriculum
01El oficio: de escribir código a dirigir agentes
- Introduction: what changes in your day when the agent arrives
- The inverted skill hierarchy
- What stays yours: judgment, design, and evaluation
- Anatomy of the daily work loop
- Task triage: what you delegate, what you pair on, and what you do yourself
- Intent before prompt: what you write before you write the prompt
- The hidden cost of a badly run loop
- Project: your log and your measured baseline
02Desarrollo dirigido por especificación
- Introduction: the specification is the new source code
- Why long prompts stop working
- Anatomy of a workable spec
- Acceptance criteria that verify themselves
- Decomposition: the right size for a task
- Specifying against existing code
- Reviewing the plan before letting it execute
- Project: spec pack for a real feature
03El harness: diseñar el entorno donde trabaja el agente
- Introduction: the agent is half the system
- Context engineering: what to load and what to hide
- Project instructions the agent actually respects
- Tools and permissions: least privilege for the agent
- The feedback loop: giving it automatic signal
- Reproducible commands, test data, and environment
- Session management: context that survives a restart
- Project: leaving a repository agent-ready
04Verificación y revisión: revisar, no aceptar
- Introduction: why 'it compiles' isn't 'it's correct'
- The trust gradient by type of change
- How to read a generated diff
- Deriving the checklist FROM your specification and turning it into a PR template
- Tests as the contract, not decoration
- API verification and version drift routine
- When to fix and when to reject and restart
- Project: run three generated changes through the procedure
05Presupuesto y control: costo, tiempo y contexto
06Gobierno del código generado por IA en equipo
- Introduction: from personal discipline to team policy
- Authorship and accountability: who signs the code
- Shared, versioned, and reviewed instructions
- PR policy and quality gates
- Architectural drift and recorded decisions
- Licensing, provenance, and data policy
- Growing juniors on a team that already works with agents
- Project: team playbook for AI-assisted code
Reviews
What students say
These reviews are from enrolled students who completed at least 50% of the course. We moderate reviews only on content grounds (spam, offensive language, personal data), never for being critical or negative.
No approved reviews yet.
Be the first to share your experience!