Specialization paths

Complete stages 1–6 of the core curriculum first. Finish production and reliability and distributed systems for the broad generalist route, or choose the relevant specialization below. These paths are not additional requirements for everyone.

Backend and distributed systems

Study system design, distributed systems, message queues, and performance engineering.

Build a workflow that tolerates duplicate delivery and partial failure. Document consistency, ordering, retry, timeout, idempotency, and recovery semantics.

Platform, cloud, and SRE

Study containerization, Docker, Kubernetes, infrastructure as code, monitoring, cloud computing, and SRE.

Deploy and operate the project with an SLO, a restore test, a failure exercise, and a documented rollback. Prefer the smallest platform that meets the requirement.

Security

Study authentication and security, network fundamentals, API design, and Linux.

Threat-model the project, implement resource-level authorization, test abuse cases, and review dependencies and deployment configuration. Verify current details against primary standards and OWASP guidance.

Data engineering

Study databases, data engineering, message queues, and cloud computing.

Build a replayable pipeline with data contracts, late-data handling, lineage, quality checks, and deletion behavior. Demonstrate recovery from a partial run.

AI engineering

Study AI, LLMs, RAG, prompt and context engineering, and MLOps. Read AI agents only when a fixed workflow is insufficient.

Add an AI feature with a non-AI baseline, representative evaluation set, cost and latency budget, authorization boundary, and fallback. Treat model and vendor comparisons as dated snapshots.