Online Judge / LeetCode System Design | System Design Interview
A user submits code written by a stranger — code that might loop forever, fork-bomb the machine, or try to read the hidden tests. Safely executing it at contest scale is the whole problem.
Problem statement Design an online judge like LeetCode: a user submits source code for a coding problem, the system runs it against a set of hidden test cases under strict limits, and returns a verdict — Accepted, Wrong Answer, Time Limit Exceeded, and so on. In scope: submitting code and receiving a graded verdict, running submissions against hidden tests under resource limits, and a contest mode with a live leaderboard during a synchronized flood of submissions. Out of scope: the in-browser code editor, problem authoring and curation, discussion forums, and the exact contest-rating…
saved by
related reading
- GitHub - ombharatiya/FAANG-Coding-Interview-Questions: A curated List of Coding Questions Asked in FAANG Interviewsgithub.com
- GitHub - donnemartin/system-design-primer: Learn how to design large-scale systems. Prep for the system design interview. Includes Anki flashcards.github.com
- [2203.07814] Competition-Level Code Generation with AlphaCodearxiv.org
- A student's dream: hacking (then fixing) Gradescope's autogradersaligrama.io
- ALE-Bench: A Benchmark for Long-Horizon Objective-Driven Algorithm Engineering | alphaXivalphaxiv.org
- NYSRGnotes.ekzhang.com
- Every Benchmark is Brokenjonathanpgabor.substack.com
- Quantifying infrastructure noise in agentic coding evals \ Anthropicanthropic.com
- FrontierCode 1.1 | Cognitioncognition.com
- Senior SWE-Benchsenior-swe-bench.snorkel.ai
- FrontierSWEfrontierswe.com
- Coderbyte | Screen, Interview & Upskillcoderbyte.com