ComingUp ComingUp
PraxisAI

PraxisAI

Practice AI-agent coding interviews on real GitHub issues, graded by tests

Oct 7, 2026 Education & Learning
ai_agent coding-interviews feedback github issues real projects

Gallery

PraxisAI

About

Practice AI coding interviews with real GitHub issues and take-home projects in a browser workspace. See your score and verdict; paid plans add detailed feedback and replay.Coding interviews increasingly let you use an AI agent, and what gets graded is how well you work with it on a real codebase. PraxisAI gives you real SWE-bench issues (Django, Flask, SymPy, matplotlib) checked out at the commit where the bug still existed, browser VS Code, an AI agent, a timer and a token budget. Hidden tests decide whether you fixed it.Free: 2 full-model rounds per week, score + verdict + test outcome, no card. Paid ($8 / $15 per month): 6-dimension report with evidence, prompt coach, token and test-cadence analysis, and full replay of prompts and agent actions. Also 24-hour take-home projects and a leaderboard.

Comments (0)

No comments yet. Be the first to comment!