What It Does
NetNerve takes your .pcap files and uses AI (LLaMA-3 via Groq) to translate raw network data into actionable security insights. Instead of staring at TCP flags and port numbers, you get plain English explanations of what's actually happening in your network traffic.
Tech Stack & Challenges
Frontend: Next.js 14 with TypeScript for the interface
Backend: FastAPI with Python/Scapy for packet processing
AI: LLaMA-3 via Groq API for analysis
Architecture: Privacy-first - files processed in memory, never stored
Try it: https://netnerve.vercel.app (supports .pcap/.cap files up to 2MB)
Looking for feedback from anyone who deals with packet analysis regularly - what would make this more useful for your workflow? Are there specific protocols or attack patterns you'd want better coverage for?