On Lounge since Oct 2026
Raghav Saravanan
Member #1,065DeveloperL2AI Engineer Intern, https://bondingai.io
AI Engineer | Starting with the customer experience. Working backwards to the technology.
How karma works. Everyone starts with 5. You get +1 each time another member upvotes one of your posts or comments (your own upvotes don't count), and lose it if they take the upvote back. Post what you're building, answer questions and help other members to earn more.
Achievements
Plugged In ×3Connected sources: GitHub, Hugging Face, LinkedIn, X, website.How: Connected 3 sources (GitHub, LinkedIn, X or a website) in Settings
Plugged In×3Featured
ShipperProjects on your profile.How: Added 1 project to their profile
Shipper
Pinned projects
Label a GitHub issue, get a review-ready pull request back.
Hands-Free AI Voice Agent for Mechanics
Testing for Issue-to-PR Autonomous Agent
Stack
About
A running changelog, roughly in order:
v0: Python + Technical Indicators — Pulled S&P 500 data, scored stocks against demand-zone criteria, shipped it as a Streamlit app.
v1: Pine Script — RSI, Fibonacci retracements, and support/resistance fused into one signal, live on TradingView.
v2: Computer Vision — Real-time pose, hand, and face landmark detection (MediaPipe + OpenCV), dynamic time warping to score how closely two motion sequences match. FastAPI backend, Supabase for data, shipped to iOS via Capacitor.
v3: Full-Stack Product — React/TypeScript frontend, Supabase for auth and data, geolocation via Mapbox, motion design with Framer Motion.
v4: RAG — Vector search over real documents (Pinecone), grounding LLM output in source material instead of letting it hallucinate with confidence. OAuth with external mail providers, Express/TypeScript backend.
v5: Agentic AI — Label a GitHub issue, get a review-ready PR: Actions trigger → AI proposes a fix → pytest verification gate with hard timeout → retry on failure (max 3) or roll back with no PR. Force-push banned, daily PR budget capped.
v6: Voice RAG — a hands-free assistant for mechanics that listens while they work, pulls torque specs and service procedures from their own shop manuals, and answers in plain English without them ever leaving the engine bay.
v7 (Present): bondingAI - Learning & Contributing
v8: ?
v9: ?
v10: 🧑🌾
Experience
Full-Stack + Applied AI Engineer | Computer Science & Business Intelligence | University of Texas at Dallas
Intern | · bondingAI
Education
University of Texas at Dallas