Design It — system design practice
7 scaffolded parts per problem: requirements, APIs & data model, architecture, core mechanism, extended features, security, and scalability tradeoffs — each graded against a rubric, not a vibe check. 3 problems, all 2 free, no card required.
Design a URL Shortener
Design a service like bit.ly — takes a long URL, returns a short code that redirects back to it.
Design a Rate Limiter
Design a rate limiter that a shared API gateway calls before forwarding each request, protecting downstream services from abusive or runaway clients.
Design an Application Performance Monitoring System
Design an APM platform like New Relic or Datadog: agents on customer hosts ship metrics, traces, and logs; the platform ingests, stores, and lets customers query and alert on them.
Want more graded practice? Code Review and Bug Hunt have 19 paid problems between them, same rubric grading as these two.
See plans →System design interview practice, answered
A system design interview asks you to scope, diagram, and defend an architecture under constraints an interviewer keeps tightening. Senior Loop's Design It problems are staged the same way a real round is run — a fixed 7-part scaffold covering requirements, APIs & data model, architecture, core mechanism, extended features, security, and scalability tradeoffs — each part graded against a rubric, not a vibe check.
What is a system design (HLD) interview?+−
You're given an open-ended prompt — design a rate limiter, a URL shortener — and asked to scope requirements, sketch an architecture, and defend tradeoffs as the interviewer pushes on scale and failure modes.
How is this different from LLD (Build It)?+−
HLD is services, data stores, and scale at the architecture level. LLD is one level down — the actual classes and interfaces that make a single component correct. Design It practices the former; Build It practices the latter.
How does grading work without a human interviewer?+−
Each of the 7 parts has a rubric — specific points a senior answer should cover, plus (for the architecture part) the components and connections a correct design needs. The AI grader checks your submission against that rubric, not against a vague quality heuristic.