backend

⚙️ Backend live

Python-first path: fundamentals to production systems — 16 modules, 5 projects, 12–18 months at 10 h/wk

▸ Part I. Foundation (months 1–4) live

  • Environment, terminal, Git: branches, pull requests, clean history
  • Python core: types, containers, comprehensions, error handling
  • OOP and idiomatic Python: dataclasses, protocols, typing
  • Testing basics: pytest fixtures and parametrization
  • Algorithms: arrays, hash tables, two pointers, complexity analysis
  • Project #1: CLI application with tests and CI

▸ Part II. Backend developer (months 4–9) live

  • Network, HTTP, Linux: methods, headers, auth, curl, DevTools
  • SQL and PostgreSQL: schema design, joins, transactions, indexes
  • Web frameworks: FastAPI and Django, API design crafts
  • Docker, CI/CD, deploy: multi-stage images, pipelines, readiness
  • Project #2: deployed REST API with authorization and PostgreSQL
  • Outcome: ready to respond to junior openings

▸ Part III. Professional engineering (months 9–15) live

  • Python internals: typing, concurrency, performance profiling
  • Architecture: ports and adapters, layered design, review culture
  • Caches, queues, events: Redis, broker patterns, idempotency, outbox
  • Middle-level testing: integration, contract-like suites, quality gates
  • Observability and operations: RED metrics, logs, incidents, postmortems
  • Backend security: OWASP, secrets management, rate limiting
  • System design: capacity, consistency, graceful degradation
  • Projects #3–4 plus capstone system

▸ Part IV. Work and growth (from month 8–9) live

  • Interviews: coding, system design, behavioral
  • First 12–24 months on the job
  • Feedback loops: interview gaps re-prioritize learning
  • Senior track: ownership, mentoring

▸ Part V. Highload System Design (10k+ RPS), Outbox & Stateful Runbooks live

  • Edge-to-Core architecture for 10,000+ RPS: GeoDNS, CDN/WAF, L4/L7 balancers, and 3-AZ stateless pods
  • Cache-Aside in Redis/Valkey with jittered TTLs and singleflight stampede protection
  • Transactional Outbox pattern for atomic PostgreSQL writes and Kafka event publishing
  • High-availability PostgreSQL (CloudNativePG, PgBouncer transaction pooling, WAL-G S3 PITR)
  • Explicit RTO (< 30s) and RPO (0 lost transactions) engineering and lock_timeout migration hygiene

Source: roadmaps/backend.json · ETA: 2026-11