Platform, Fintech & AI: Robinhood-style Brokerage, CI/CD Runner (GitHub Actions), Online Chess Server, Calendar & Scheduling (Google Calendar / Calendly) & RAG Document Q&A

Five designs that interviewers moved into their loops in 2025–2026: a retail brokerage that must stay correct at the opening bell, a CI platform that runs strangers' code fairly and safely, a real-time game server where 300 ms decides a game, a calendar where time itself is the hard part, and an enterprise "chat with your docs" system where retrieval, not the LLM, decides quality.

About 200 minutes · 20 exam questions · 14 interview questions · 2 code labs

What you will learn

  1. Robinhood-style brokerage (1/3): requirements, estimation, API & data model
  2. Brokerage (2/3): order lifecycle — pre-trade risk, outbox, smart order routing over FIX
  3. Brokerage (3/3): quote fan-out, surviving the opening bell, positions vs settlement, and trade-offs
  4. CI/CD runner platform (GitHub Actions) (1/3): requirements, estimation, workflow model & API
  5. CI/CD runner (2/3): architecture — DAG orchestrator, fair-share dispatch, warm ephemeral runner pools
  6. CI/CD runner (3/3): isolation & secrets, live log streaming, artifacts & caching, trade-offs
  7. Online chess / multiplayer game server (1/3): requirements, estimation & rating-based matchmaking
  8. Online chess (2/3): the authoritative game server — move protocol, server clocks & reconnection
  9. Online chess (3/3): spectators, failover, cheating detection and trade-offs
  10. Calendar & scheduling (Google Calendar / Calendly) (1/3): requirements, estimation, API & data model
  11. Calendar (2/3): RRULE expansion, time zones & DST, and free/busy queries
  12. Calendar (3/3): invitations, sync, reminders at scale, Calendly booking and trade-offs
  13. RAG document Q&A ("chat with your docs") (1/3): requirements, estimation & the ingestion pipeline
  14. RAG (2/3): vector index sizing (HNSW vs IVF-PQ), hybrid retrieval, reranking & permission-aware search
  15. RAG (3/3): grounded generation, evaluation, caching, cost/latency — and wrap-up across the five designs
  16. Level bar: what mid, senior and staff answers include
  17. Deep-dive ladders: bad → good → great

Previous: Consumer & Marketplace: Flash Sale (Big Billion Days), Price Tracker & Alerts (CamelCamelCamel), Tinder, News Aggregator (Google News) & Strava · Next: Instagram & Food Delivery (Swiggy / Zomato / DoorDash)