Hey developers! I built ResumeTrust AI to solve the issue of exaggerated resumes by cross-referencing candidate skills with their live GitHub activity natively.
⚙️ Key Backend Optimizations:
24-Hour Cache: Uses findOneAndUpdate to cache results and save Gemini API credits.
Data Privacy: In-memory PDF parsing with immediate auto-deletion from the server.
Security Gate: Backend-enforced 403 limits for candidate accounts.
🌐 Live Link: [[https://resume-trust-ai-wccy.vercel.app](https://resume-trust-ai-wccy.vercel.app)]
💻 Codebase & README: [[https://github.com/Vinay-pandey-coder/ResumeTrust-ai](https://github.com/Vinay-pandey-coder/ResumeTrust-ai)]
👇 If you find my code structure and logic helpful, please drop a ⭐ on the GitHub repository to support my portfolio!
source & further reading
dev.to — original article
I tracked Claude Code and Codex pass-rates for 95 days — what "getting dumber" actually looks like
How I Built an AI Agent That Earns $500/Month in Open Source Bounties — Full Architecture, Real Code, and Honest Numbers After 72 Hours
Yo Vibe Coders, Stop Building FE Slop and Use Telegram Groups Instead