Why this handbook exists.

Between 2023 and 2026, three forces reshaped what engineers are expected to know.

The syntax layer collapsed. AI coding assistants—Copilot, Cursor, Claude Code—handle the keystrokes. New grads now ship production code on day one. The differentiator moved up the stack: from writing a loop to designing a system that survives Black Friday.

The interview bar moved. System design rounds now appear at levels that rarely saw them five years ago—the mid-level bar, not just the senior one. What used to be a senior-only round is now routine at SDE 2 and even SDE 1.

The stack itself shifted. CRDTs, vector search, LLM serving, and RAG pipelines were niche in 2022; they are table stakes in 2026. Chat systems include presence and end-to-end encryption by default. Payment platforms include idempotency and event sourcing by default. The vocabulary expected to discuss them has followed.

The gap between “can code” and “can design” used to be a four-year apprenticeship. Now it is the gate—and it opens earlier.

The open-source system design landscape is full of well-intentioned projects that share a common failure mode. Some are link dumps — curated READMEs of external URLs to blog posts of uneven quality. Some are teaser-and-redirect — the README hints at substance, then steers you toward a paid course on another site. Some are frozen in 2022 — no LLM serving, no RAG, no vector search, no CRDT-based collaboration. And some are monolithic READMEs — a single 100 KB file with no ordering, no diagrams, and no way to track what you have learned.

This handbook is none of those things.

Written as inline Markdown in a public Git repository. Rendered by a static site generator. Every chapter is peer-reviewable, forkable, and translatable. Nothing is paywalled. Every external citation is from a primary source — papers, engineering postmortems, official docs — not blog posts about blog posts.

Every chapter follows the same arc. A mental model gives you intuition — the shape of the problem before the mechanics. An architecture section adds the diagrams, the primitives, and the concrete systems. A trade-offs analysis names what each choice costs and when to make it. A further-reading list closes the loop with primary sources you can audit yourself.

Trade-off guides follow a stricter 13-section template so you can navigate them consistently: TL;DR, learning objectives, the core tension, side-by-side comparison, when to pick each side, the hybrid path, real-world examples, common mistakes, decision checklist, key takeaways, further reading, and flashcards.

This project stands on the shoulders of others. We do not copy from any of them. We teach the same concepts in our own words, with our own diagrams, and we cite them.