I didn't know. I thought I solved everything more or less equally. And I would've kept not knowing the truth if I hadn't built a tool to analyze it!

BetterForces is like Codeforces Profile, but with numbers and charts that actually help you train.

What I learned about myself (as example): I have 23 abandoned problems at rating 2400+. I tried, got WA 5 times, gave up. Now I have a list — can approach them systematically. My average rating on tree problems is higher than overall. That's my strength. Constructive algorithms? Disaster. Need to practice that.

Works for any handle: https://betterforces.deyna.xyz/

Open source, if you want to check it out or contribute: GitHub.

Star the repo to support me!

What metrics would you like to see in your profile?

Have you ever wanted to automatically retrieve problem editorials from the archive?

Codeforces Editorial Finder is an open-source HTTP API that uses GPT-4o to automatically find and extract clean solution text, complexity, and hints.

Key Features: AI-Extraction: Automatically parses blog posts and contest pages. Format Support: Handles dynamic content (JS) and PDF tutorials. Performance: Built with LiteStar, Playwright, and Redis.

Contributors Wanted! The project uses Python 3.13, uv, and Ruff. We need help with unit tests, prompt tuning, and security!

GitHub: https://github.com/deyna256/codeforces-editorial-finder