Phases & Labs

This curriculum has 7 phases and 46 labs. Phases 1–5 build a database and the consensus protocols underneath it; Phase 6 turns to the adjacent domain of cloud gateways and application networking (the Netflix Cloud Gateway track); Phase 7 zooms out to platform & distributed-systems architecture (the Apple Software Architect track) — how you decompose, connect, partition, ship, and govern a platform. Phases build on each other, but within Phase 4 (consensus) you can do Raft → Paxos → ZAB in any order after the foundations in db-16, and Phases 6 and 7 can each be read independently of Phases 2–5 once you have the distributed-systems mindset from Phase 4.

Legend: ✅ complete · 🟡 scaffolded · ⬜ planned

Phase 1 — Storage Primitives & Foundations

Before you can build a database, you need to understand the medium it lives on.

Phase 2 — LevelDB / LSM-Tree

Build a production-shape LSM-tree key-value store, the way Google built LevelDB and Meta forked it into RocksDB.

Phase 3 — SQLite / B-Tree

Build a B+-tree storage engine, a pager, a SQL parser, a bytecode VM, and a transaction manager.

Phase 4 — Consensus Algorithms

The three canonical consensus families — implemented, tested, and compared.

Phase 5 — Advanced Storage & Capstone

Phase 6 — Cloud Gateway & Application Networking

The adjacent domain: the L4/L7 data plane, the API gateway, the WebSocket fleet, the control plane that programs them, and the Kubernetes substrate they run on. Built toward the Netflix Cloud Gateway role; each lab maps to a JD requirement and a Netflix talk. See gw-00 overview and the interview playbook.

Phase 6 note on shape: Phases 1–5 prove correctness with byte-identical cross-language dumps. Networking systems aren't byte-deterministic, so Phase 6 proves things the industry way: real, compilable, go test -race-green mini-implementations (stdlib-only Go, offline) plus demo programs that print each lab's headline result. Each lab ships CONCEPTS.md (the why), a maintainer- level GUIDE.md (the deep, hands-on companion), references.md, docs/ (analysis + execution + verification), code-rich steps/, scripts/verify.sh, and a working src/go/ you hack on. Run the whole phase with bash gw-00-cloud-gateway-overview/verify-all.sh. Start with the Hitchhiker's Guide.

Phase 7 — Platform & Distributed Systems Architecture

The architect's view: how you decompose a platform into services, connect them (sync + async), partition and keep data consistent, ship it (IaC, GitOps), keep it reliable (SLOs), and govern the decisions. Built toward the Apple Software Architect – Distributed Systems & Platform Engineering role; overlapping topics (operators/CRDs, service mesh, tracing, circuit breakers) are cross-referenced to Phase 6 / the consensus phase rather than rebuilt. See pa-00 overview, Hitchhiker's Guide, and the architect interview playbook.

Phase 7 shape: same as Phase 6 — real, go test -race-green, stdlib-only Go (an event bus, a partitioned log, the outbox/saga, a consistent-hashing ring, an IaC engine, a GitOps reconciler, an SLO engine, fitness functions) + a maintainer-level GUIDE.md per lab. Run it all with bash pa-00-platform-architecture-overview/verify-all.sh.

Suggested Pace

• Full-time learner: ~2 labs per week ⇒ ~12 weeks end-to-end.
• Side-project learner: ~1 lab every 1–2 weeks ⇒ ~6 months.
• Reading-only path: skim CONCEPTS.md + docs/analysis.md per lab ⇒ ~1 week for the entire curriculum.

Recommended Progression

Phase 1 (must do all 4 in order)
   │
   ├─→ Phase 2 (LevelDB)  ──┐
   │                        │
   └─→ Phase 3 (SQLite) ────┤
                            ↓
                         Phase 4 (Consensus) ──→ Phase 6 (Cloud Gateway)
                            ↓
                         Phase 5 (Capstone)

Phase 2 and Phase 3 are independent — pick the storage style that excites you first. Phase 4 only references Phase 1 fundamentals, so you can detour into consensus early if you want. Phase 5's capstone assumes all four prior phases.

Phase 6 is a standalone track. It targets the Netflix Cloud Gateway role and only assumes the distributed-systems mindset from Phase 4 (the data-plane/control-plane split echoes consensus; config propagation is a consistency problem). If you're prepping for an application-networking / gateway role, you can jump straight to gw-00 after Phase 4. Within Phase 6, do gw-01 → gw-03 in order; after that the branches are independent and gw-12 (capstone) assumes the rest.

Phase 7 is also a standalone track, aimed at the Apple Software Architect role. It assumes only the Phase 4 distributed-systems mindset and cross-references Phase 6 for the overlapping infrastructure topics (operators/CRDs, service mesh, tracing, circuit breakers). Within Phase 7, do pa-01 → pa-05 in order (the comms + data spine); after that pa-06…pa-09 are independent and pa-10 (architecture-in-practice) ties it together. Start at pa-00.