ai-platform/roadmap
All pages

Start here

  1. How to use this roadmap
  2. Two kinds of role
  3. Read code, write it rarely
  4. Target profile
  5. Eight principles
  6. Work on a real product

Phases

  1. 01Direct coding agents and read their code
  2. 02Learn how models behave and what they cost
  3. 03Build retrieval that finds the right evidence
  4. 04Build tool servers with MCP
  5. 05Design agent workflows that survive failure
  6. 06Run agent workloads on AWS
  7. 07Trace every step an agent takes
  8. 08Build evals that catch regressions
  9. 09Run agents reliably in production
  10. 10Secure agents and the code they write
  11. 11Build an internal AI platform
  12. 12Get teams to adopt the platform
  13. 13Build Forge, the capstone platform
  14. Elective: Kubernetes

After the phases

  1. Staff skills
  2. 12-month plan
  3. A typical week
  4. Show your work
  5. Tools to know
  6. What to skip
  7. Final checklist

Elective · Only if a target role needs it

Run agents on Kubernetes

The roles this roadmap targets rarely ask for Kubernetes. The teams that hire for them usually run serverless and managed services on AWS. A full phase on Kubernetes would take time away from evals, security, and adoption, so it is an elective. Take it if you target roles that run long-lived or high-volume agent workloads on Amazon Elastic Kubernetes Service (EKS).

If you take it, learn pods, deployments, services, jobs, config maps, secrets, resource requests and limits, probes, network policies, and Helm. Learn to autoscale on queue depth instead of CPU. Then learn how identity works on EKS. Run the Phase 6 worker locally, on ECS, and on Kubernetes, with state stored outside the worker, graceful shutdown, and heartbeats. Write down when a workload does not need Kubernetes. Knowing when to say no to Kubernetes is also a senior skill.