Common questions about OSBridge, matching, privacy, and contributing.
5 min readOSBridge is an AI-powered platform that helps developers transition from solving algorithmic coding problems on LeetCode to making meaningful contributions to real open-source projects. It analyzes your problem-solving patterns and matches you with open-source issues that align with your skills.
Yes, OSBridge is completely free to use. There are no premium tiers, paywalls, or hidden costs.
OSBridge is designed for students with strong DSA skills but little project experience, developers who have never contributed to open source, GSoC aspirants, hackathon participants, and anyone who finds GitHub overwhelming.
Match scores are based on cosine similarity between your skill embedding and the issue embedding. They indicate how closely your algorithmic strengths align with the issue requirements. Scores above 70% are strong matches, 40-69% are moderate, and below 40% are filtered out.
Low scores usually mean your LeetCode profile needs more diversity. Try solving problems across different tag categories (not just Arrays and Strings). The more diverse your solved tags, the richer your skill embedding becomes.
Yes! OSBridge offers alternative profiling methods: GitHub profile analysis, resume upload, and a Vibe Check quiz. However, LeetCode syncing provides the most accurate skill profiling.
Issues are synchronized from curated repositories periodically by admins. Stale issues (older than 90 days, with linked PRs, or with high comment counts) are automatically filtered out.
We collect your public GitHub profile (username, avatar) via OAuth, and your LeetCode solved tag counts via their public API. We never access private repositories, email addresses, or personal information beyond what's publicly available.
Yes. Contact us and we will remove all your data from our systems including skill profiles, match history, and feedback records.
OSBridge uses Google Gemini's text-embedding-004 model for generating 768-dimensional vector embeddings and Gemini 2.5 Flash for generating personalized 'Why You' match explanations.
LeetCode profile data is cached in Upstash Redis with a 24-hour TTL. This prevents excessive API calls to LeetCode's servers and speeds up subsequent profile lookups.