Why This Role Matters
JustCall is the communication layer for tens of thousands of revenue teams - calls, SMS, and increasingly AI voice agents that talk to customers in real time. We've earned a decade of scale. We've also inherited a decade of architecture: a large PHP/JavaScript surface area, an integrations layer that's hard to extend, and a voice stack built for a pre-LLM world.
We are not looking for people to staff a backlog. We are looking for a handful of engineers who have already built systems at scale, who can walk into a ten-year-old codebase and a green-field voice-agent problem in the same week, and who want the rare thing: full ownership of a hard problem, with the autonomy and the equity to match.
If you read the four problems below and feel the itch to start sketching the design, this is for you.
Tracks
Every Principal Engineer shares the bar above. Tracks describe where you'll start and the depth we'll index on. Strong candidates often blur two tracks - that's welcome.
Track A - Voice & Agent Platform Principal Engineer
Your charter.
Own the architecture of JustCall's next-generation voice agents end-to-end, from the telephony edge to the model orchestration to the evaluation harness that tells us whether an agent is actually positive.
What you'll own in your first 6–12 months
- A streaming, cascaded voice pipeline (ASR → LLM → TTS) with end-to-end response latency under ~800ms and a credible path toward the ~500-600ms range where callers stop noticing they're talking to AI.
- Turn-taking and barge-in that feel human - semantic endpointing, interruption handling, and full-duplex audio - not naive VAD silence timers.
- An evaluation and observability stack for voice: latency budgets per stage, transcription, hallucination and instruction-following evals, and replayable call traces.
- The orchestration layer: tool/function calling, RAG over customer knowledge, guardrails, and graceful fallback when a model or vendor degrades.