Pallav Burnwal

Pallav Burnwal

Full-Stack Developer & AI Engineer

About

Full-stack developer building scalable web applications, real-time media systems, and AI-powered tools. Experienced across React, TypeScript, Next.js, Python (FastAPI), and cloud infrastructure. Always exploring new developer tools, workflow automation, and shipping fast, user-focused products.

Badges

Tastemaker
Tastemaker
Gone streaking
Gone streaking
Gone streaking 5
Gone streaking 5

Maker History

  • Uvilox AI
    Uvilox AIReal-time AI sign language interpretation & emergency care.
  • 🎉
    Joined Product HuntMarch 1st, 2026

Forums

•

10h ago

Building Uvilox AI: Real-time sign language & voice translation in the browser

Hi everyone! I ve been working on Uvilox AI, a browser-native communication platform designed to bridge real-time conversations between Deaf/hard-of-hearing users and hearing speakers.

What it does & how it works:

  • Real-time Sign Language Recognition: Uses computer vision hand/pose landmark tracking to translate sign language gestures into text and natural-sounding speech on the fly.

  • Two-Way Live Video & Voice: Built on WebRTC (LiveKit) and low-latency speech-to-text/text-to-speech so both participants can communicate naturally without switching apps.

  • Zero-install web access: Designed to run directly in modern browsers alongside a companion Android app.

Where I d love your feedback: Since accessibility tools live or die by speed and clarity in real-world conditions (varying lighting, camera angles, and network speeds), I m currently focused on refining the onboarding flow and latency indicators.

•

10h ago

Engineering trade-offs in sub-second browser video + AI inference pipelines

Hey Makers,

Over the last few months, I ve been building a real-time accessibility web app that runs computer vision landmark detection alongside live WebRTC audio/video streams directly in the browser.

One of the biggest challenges has been keeping end-to-end latency low enough for natural, real-time conversation without melting the user's CPU or hitting heavy cloud GPU costs. A few architectural lessons that worked well for us:

  • Client-side landmark extraction: Running lightweight pose/hand landmark tracking locally in the browser and only streaming compact coordinate tensors over the network cut bandwidth and inference latency dramatically compared to streaming raw video frames.

  • Decoupling media from inference: Keeping WebRTC media routing separate from the AI processing pipeline prevented audio jitter whenever model inference spiked.

  • Graceful degradation on lower-end devices: Dynamically adjusting frame-sampling rates based on client FPS kept the experience usable on budget hardware.

We were #1 on Product Hunt. Here's what we've been building since!

Back in August, NudgeForMe became #1 Product of the Day on Product Hunt.

First, a big thank you to everyone here who supported our launch!

View more