Interview prep that doesn't want your subscription.
DesignDojo is a free, open-source tool for practicing system design (HLD) and low-level design (LLD) interviews. You bring your own AI key — or run a local model — and an interview-grade tutor walks you through real problems stage-by-stage.
Why this exists
Most interview-prep platforms charge $30–$60 a month and tightly couple practice with their own AI. That model is expensive, and you're paying for AI usage you may already have access to. DesignDojo separates the two. The app is free. AI is your choice — OpenAI, Anthropic, Gemini, OpenRouter, Groq, or a local model on Ollama. A full interview session typically costs a few hundredths of a cent in tokens.
The point isn't cheap — it's honest. You shouldn't need to subscribe to learn how a URL shortener scales.
Your AI key, stored only in your browser. We never see it.
Point at Ollama or LM Studio and your data never leaves the laptop.
MIT licensed. Read the code, contribute problems, fork it.
What you get
- 30+ interview problems across HLD (URL shorteners, news feeds, ride-sharing, ad aggregators) and LLD (parking lots, chess, vending machines, file systems).
- Senior-level write-ups for each problem — what a real interviewer is listening for, where mid-level answers fall short, what staff+ depth looks like.
- Stage-by-stage practice mode with a built-in whiteboard (Excalidraw) for HLD or a Monaco editor for LLD.
- AI tutor that grades each stage against the same rubric senior interviewers use, returning specific must-fix gaps instead of generic praise.
Who built this
DesignDojo is built and maintained by Aditya Jindal. Issues, PRs, and new problem suggestions are welcome on GitHub.