Intervu.dev was created to close a gap that most practice tools ignore. LeetCode-style platforms teach you to produce correct solutions, but real technical interviews test something different: can you explain your reasoning, defend your trade-offs, and write working code while someone watches and asks follow-up questions? Intervu.dev reproduces that pressure in a structured AI interview loop.
Each session follows the flow a real FAANG interview uses — clarify requirements, design an approach, code live, then test and iterate. The AI interviewer asks probing follow-ups rather than simply grading output, and afterwards you receive signal-based feedback on clarity of communication, quality of trade-offs, and code structure. You can run sessions in Python, JavaScript, Java, C++, and other major languages, choosing language and difficulty before you begin.
You can either start a full FAANG-style mock interview or bring a specific LeetCode problem in and practice it as an interview. The site also publishes a blog and maintains open source components, and offers a way to try the experience free before committing to paid practice.
