lb.

Track 02 · In development

The fundamentals that outlast any framework.

System design from first principles: scale, storage, queues, and failure, in production terms.

System design taught properly: estimation, networking, databases at scale, caching, queues, reliability, and case studies. Start from first principles, leave with production instincts and the numbers to defend your designs.

One email when module 1 lands. No spam; reading stays free, no account.

learnbackend.com/system-design · the syllabus, as a system
clientload balancer02app servers
cache04database03
queue05workers

Also on the syllabus

The track, drawn as the system you will learn to design. Every box is one module, done properly; tap one to jump to it.

07+modules planned
40+lessons queued
$0forever
01email when it ships

Planned curriculum

Seven modules, in the order you'd build the system.

  1. Foundations and Estimation

    up next

    The math that settles design arguments before they start: back-of-envelope estimation, the latency numbers every engineer should know, SLOs, and capacity planning.

    napkin mathlatency numbersSLOscapacity planning

  2. Networking and API Design

    in queue

    What actually happens between the client and your handler: load balancers, TLS, rate limiting, and the API decisions you cannot walk back after launch.

    load balancersrate limitingTLSversioning

  3. Databases at Scale

    in queue

    What breaks first when one box stops being enough: replication, partitioning, consistency models, and the indexing choices that decide your read path.

    replicationshardingconsistencyindexes

  4. Caching

    in queue

    Cache-aside to CDNs: where each cache layer belongs, how to invalidate without serving lies, and when a cache hides a design problem instead of fixing one.

    cache-asideinvalidationCDNsTTLs

  5. Queues and Async Processing

    in queue

    Moving work off the request path without losing it: delivery guarantees, the outbox pattern, backpressure, and when a queue beats a cron job.

    delivery guaranteesoutboxbackpressurestreams

  6. Reliability and Resilience

    in queue

    The difference between an incident and an outage: retries, timeouts, circuit breakers, idempotency, and graceful degradation under real failure.

    retriestimeoutscircuit breakersidempotency

  7. Case Studies

    in queue

    The whole map, end to end: design a URL shortener, a news feed, a chat system, and a rate limiter, with the numbers to defend every choice.

    url shortenernews feedchatrate limiter

While you wait

Already live and close in spirit: reliability and networking for production systems.

Be first in when module 1 lands

One email at launch, then lessons as they ship. Free either way; email is the ping, not the gate.

Meanwhile, everything in the AI Engineering track is live now.