Reqflow
← All learning paths
Learning path
Advanced~93 min·4 systems

Big Tech Systems Explained

The architectures behind Netflix, TikTok, Airbnb, and the stock exchange — each one a masterclass in a different hard problem at scale.

After this path you will be able to

Understand how the world's most-studied production systems actually work — and use them as reference architectures when an interviewer asks you to design 'something like Netflix' or 'a marketplace like Airbnb'.

Interview approach for this path
  1. 1.Use the real system as a reference architecture, but make sure you can explain why each design decision was made, not just what it is.
  2. 2.Identify the one hard problem that makes this system interesting: CDN edge caching for Netflix, fan-out at Twitter scale, marketplace double-booking at Airbnb, microsecond matching at the exchange.
  3. 3.Explain the bottleneck tier and how the real system solved it. Interviewers want depth on the hardest part, not breadth across the whole thing.
  4. 4.Address scale numbers honestly. Netflix serves 200M subscribers. The exchange processes millions of orders per second. Make sure your design math supports those numbers.
  5. 5.Call out the trade-offs the real system made and whether you agree. 'Netflix chose to precompute recommendations offline and accept some staleness, which makes sense because freshness matters less than personalization quality.'
  6. 6.Finish with what you'd do differently if you were starting from scratch today versus when the system was originally built.

Systems in this path

4 total
  1. 1
    Netflix (Video Streaming)
    Advanced·25 min

    CDN-first, adaptive bitrate, recommendation.

  2. 2
    TikTok (For You Feed)
    Advanced·25 min

    Video pipeline, ML-driven recommendation, infinite scroll.

  3. 3
    Airbnb (Marketplace)
    Intermediate·18 min

    Geo-search, atomic seat-holds, authorize-then-capture payments, async fan-out.

  4. 4
    Stock Exchange (Matching Engine)
    Advanced·25 min

    Single-threaded per-symbol matching, in-memory order book, microsecond latency.

Concepts reinforced throughout