Leads the team this role joins
Hires for this team
On the team, can refer you
From your own Gmail and LinkedIn.
About the role
This is a full-time founding engineer role in Gurugram, India, for an engineer with 3-5 years of experience who has run a voice AI system in production. The role sits at the intelligence layer of ixigo's self-healing voice agents for enterprise customer support, owning the pipeline that carries a phone call from speech to response. The engineer works directly against live production call traces and a traffic testing setup, and decisions ship to real calls within days.
What you’ll do
- Own the latency budget across speech-to-text, the language model and text-to-speech, aiming for under one second end to end.
- Find and fix bottlenecks that span several pipeline stages at once, such as prompt size, cache hit rate and throughput contention.
- Evaluate speech, language and voice providers, including in-region options in India, and build an abstraction that keeps the stack swappable.
- Design turn detection, barge-in handling, filler audio and retry logic.
- Add tracing across the full call lifecycle so latency regressions can be diagnosed from data.
- Decide which pipeline components to build and which to buy as the product grows.
What they’re looking for
- 3-5 years building and operating production systems, with real on-call or ownership experience.
- Hands-on production work optimizing a cascaded voice pipeline that combines speech-to-text, LLM orchestration and text-to-speech, not only a hosted voice API.
- Clear understanding of where latency hides in each stage and the accuracy-versus-speed trade-offs made.
- Solid systems fundamentals: concurrency, network I/O, profiling, and measuring before optimizing.
- Comfort with ambiguous problems and real production data rather than clean specifications.
- Direct communication and willingness to push back with data.
Nice to have
- Experience with WebRTC, SIP or PSTN telephony, or real-time audio streaming.
- Familiarity with prompt caching and how prefill cost scales with context length.
- Experience in a founding or very early-stage company that ran a system with no safety net.
What’s on offer
- Full-time role based in Gurugram.
Skills
Questions about this role
How many years of experience does this role need?
The role asks for 3-5 years of experience building and operating production systems.
Where is this job based?
The role is listed in Gurugram, India, as a full-time position.
Is voice AI experience required?
Yes. Candidates need hands-on experience with a cascaded voice pipeline, not only a hosted voice API.