SDE 4

Join a fast‑growing team tackling the toughest real‑time AI voice challenges. You’ll design, build, and scale streaming pipelines that power natural, responsive voice agents for millions of users. What You'll Do Design and scale real‑time ASR, LLM, and TTS pipelines for low latency. Implement voice activity detection, turn‑taking, endpointing, and interruption handling. Build audio streaming services using WebRTC, WebSockets, SIP, and media servers. Orchestrate AI agents with LLM integration, tool calling, and context management. Optimize end‑to‑end latency, audio quality, and infrastructure costs. Create monitoring, observability, and fault‑tolerance for highly concurrent services. What You Need 8+ years of software engineering with production real‑time or Voice AI systems. Strong Python or Go skills, plus deep async programming and concurrency expertise. Experience building streaming systems via WebSockets, WebRTC, or gRPC. Hands‑on work integrating and optimizing streaming ASR, LLM, and TTS providers. Deep knowledge of voice fundamentals: VAD, endpointing, turn‑taking, interruption handling. Familiarity with SIP, RTP, and debugging real‑time media issues. Proficiency with AWS/GCP, Docker, Kubernetes, and observability tools. Good to Have Built production Voice AI agents or conversational AI products. Used frameworks such as LiveKit, Pipecat, or Vocode. Worked with telephony CPaaS platforms, SIP infrastructure, or media servers. Experience in speech processing, audio DSP, noise suppression, or speaker diarization. The Opportunity You’ll own systems end‑to‑end in a small, impact‑driven team, with market‑benchmarked fixed pay, a variable component, and equity as the company scales toward $100 M ARR.