Saurabh R.

Jai Shri Shyam! 🙏 I architect and build real-time video, voice, and AI systems, with 12+ years of experience delivering production software. My core expertise is WebRTC, LiveKit, real-time communication, and AI-powered applications. I work with startups and established teams to build new products, improve existing systems, and solve complex engineering problems involving latency, scalability, reliability, and infrastructure. I don't just integrate SDKs. I work across the entire architecture, from signaling and media servers to backend services, AI pipelines, frontend applications, and cloud infrastructure. What I specialize in: WebRTC & LiveKit Video conferencing, audio/video calls, screen sharing, recording, live streaming, spatial audio, real-time collaboration, and custom communication platforms. AI Voice Agents & Conversational AI Real-time voice agents using LiveKit Agents, OpenAI, Claude, Gemini, Deepgram, ElevenLabs, and other STT/TTS providers. Function calling, tool execution, external API integrations, conversational memory, interruption handling, turn detection, and latency optimization. AI Agents & Automation AI agents that retrieve information, interact with external systems, execute tools, call APIs, and automate workflows. LLM integrations, RAG, vector databases, multi-step tool calling, and agent orchestration. Real-Time Video & Audio AI Live transcription, speech translation, AI meeting assistants, real-time audio processing, video analysis, and intelligent features for communication platforms. Infrastructure & Scaling Self-hosted LiveKit, SFU/TURN configuration, WebRTC troubleshooting, network debugging, performance optimization, and migrations from Twilio, Agora, Daily, or Vonage. SaaS & Full-Stack Architecture Complete SaaS platforms, backend architecture, APIs, authentication, databases, real-time messaging, frontend development, and cloud infrastructure. Technical Stack: Real-Time: WebRTC, LiveKit, Mediasoup, Janus, Socket IO, FFmpeg, GStreamer Backend: C#, Microsoft dotnet, Golang, Python, NodeJS, TypeScript, Java Frontend: React, NextJS, Angular, Electron AI & Agents: OpenAI, Claude, Gemini, LiveKit Agents, LangChain, LlamaIndex, RAG, Function Calling, Tool Execution, STT/TTS Cloud: AWS, GCP, Docker, Kubernetes, Terraform, CI/CD Databases: PostgreSQL, MySQL, MongoDB, Redis, Vector Databases How I approach projects: Before writing code, I want to understand what's actually happening and what needs to improve. If your WebRTC calls are dropping, I'll investigate the signaling, ICE negotiation, TURN servers, and network conditions. If your AI voice agent is slow, I'll look at the complete pipeline, from speech detection and transcription to LLM processing, tool execution, and audio generation. If your infrastructure costs are increasing, I'll review the architecture and identify practical ways to optimize it. I'm comfortable working with existing engineering teams, taking ownership of a specific technical challenge, or architecting and building a product from scratch. My focus is simple: build software that works reliably in production, performs well under real-world conditions, and remains maintainable as the product grows. If you're building a real-time communication platform, AI voice agent, or complex SaaS application, tell me what you're working on and where you need help. We can discuss the architecture, challenges, and best way to approach it.