Vaishnav Pillai

I’m a Software Engineer with 2+ years of experience building AI-driven, production-grade full-stack products, with deep exposure to ML/R&D and applied GenAI systems. I focus on turning applied ML research into scalable, reliable production systems.

I’ve shipped real-time speech-to-speech and multimodal AI pipelines used in live, student-facing products—combining RAG, LLMs, STT, and TTS end to end. In my Software Engineer role at 47Billion Information Private Limited, I designed and built a production-grade AI Mock Interview system around these components.

Beyond models, I engineered low-latency real-time audio/data flow using WebRTC and WebSockets to ensure stable performance at production scale. I owned the system from prototype to production, handling reliability, latency optimization, and scalability for live user traffic.

Earlier, I led a Text-to-Lips POC pipeline (speech → STT → LLM response → TTS → Wav2Lip), exploring multiple lip-sync approaches to balance output quality and inference speed for near real-time generation. My ML internship experience included building an EasyOCR-based character recognizer and an object detection system using advanced computer vision techniques—strengthening my ability to evaluate, optimize, and deploy practical ML solutions.