TechPrep

Track 2

System design

A method that scales from Senior to Principal, the trade-offs every design turns on, and worked prompts including LLM-era systems.

Design rounds grade judgement, not diagrams. The same prompt is answered differently at each level: a Senior candidate produces a working design, a Staff candidate makes the trade-offs explicit and drives the conversation, and a Principal candidate frames the problem in terms of the organisation, the migration and the risk.

Do the method course first, then the trade-offs, then the worked prompts with a timer. The mock rounds track has a scored design mock.

The design method

What each level is graded on, estimation that earns its time, and a map of the building blocks.

  1. 01How design rounds are scored, by levelThe same prompt, three bars: Senior builds it, Staff chooses between ways to build it, Principal decides whether and how the organisation should.
  2. 02Requirements and estimationTurn a vague prompt into numbers that decide the design: peak load, storage growth and fan-out, each with its uncertainty.
  3. 03The building blocks and when to reach for eachA map of storage, caching, messaging and compute choices, with the question that picks between them.

The trade-offs

Replication and partitioning, caching, asynchronous processing and reliability: the four areas where designs are won or lost.

  1. 01Replication, partitioning and consistencyCopies buy availability and read scale, partitions buy write scale, and both cost you some form of consistency.
  2. 02Caching and its failure modesCaches are the cheapest latency win and the most common source of subtle bugs: plan invalidation, stampedes and cold starts up front.
  3. 03Queues, streams and idempotencyAsynchronous work is at-least-once in practice, so every consumer must be idempotent and every pipeline needs a poison-message plan.
  4. 04Reliability: SLOs, timeouts, retries and regionsDecide how reliable the system must be, then spend that error budget deliberately with timeouts, retries, isolation and graceful degradation.

Worked prompts

Four prompts from Senior to Principal: an API rate limiter, a multi-tenant LLM gateway, enterprise document Q&A and a monolith migration strategy.

  1. 01Worked prompt: a rate limiter for a public APIA Senior-level classic, answered with Staff-level trade-offs: algorithm choice, distributed counting and failure behaviour.
  2. 02Worked prompt: a multi-tenant LLM gatewayA Staff-level, AI-era design: routing, quotas, streaming, cost attribution, safety hooks and provider failure in one service.
  3. 03Worked prompt: enterprise document Q&ARetrieval-augmented generation at enterprise scale, where permissions, freshness and evaluation are harder than the model call.
  4. 04Worked prompt: a Principal-level migration strategyMove a revenue-critical monolith to a target architecture across many teams without a freeze: the design is the plan.