Full Stack Engineer

We're hiring a Full Stack Engineering Intern at VoxMith in Noida. If you're looking for a predictable role with well-defined tasks, this probably isn't it. Voice AI in production is still incredibly difficult. Agents fail silently, debugging is often guesswork, and most teams don't have a good way to understand why conversations succeed or fail. We're building VoxMith to solve that. Our goal is to build the reliability layer for voice and conversational AI, where every failure can be traced, measured, and improved. This is a true zero-to-one problem. There isn't a playbook to follow. You'll get to work on things like: • Real-time voice systems using VoIP, WebRTC, and streaming pipelines • Observability across ASR, LLM, and TTS stacks • Converting messy conversations into structured evaluation signals • Building systems where latency, correctness, and reliability all matter We're looking for someone who enjoys solving hard engineering problems, not just assembling existing tools. You'll be a great fit if you: • Have built with LiveKit, Pipecat, Twilio, or similar voice AI frameworks, preferably beyond simple demos • Can debug across infrastructure, backend, frontend, and real-time systems • Are comfortable working without detailed specifications • Care about building correct, reliable systems instead of just shipping features quickly Bonus if you've worked with VoIP, real-time infrastructure, AI systems, or developer tooling. 📍 Location: Noida If this sounds exciting, I'd love to hear from you. Reach out here or email me at team@voxmith.com. And if reading this makes you think, "That sounds really difficult," you're probably right. That's exactly why we're building it.