About the company

Our client is an early-stage, YC-backed fintech AI company in New York automating late-stage consumer debt recovery with voice AI agents — asset research, pre-legal outreach, litigation, credit reporting, and enforcement, all agent-driven. The team is around 8 people, mostly engineers, founded in 2024 and growing fast, with a target of roughly 4x revenue growth over the next year. It's an office-first team in Flatiron/NoMad that ships quickly and expects the same.

The role

This is the seat that owns the company's entire in-house voice AI infrastructure. Today the voice agents run on a third-party provider; you'd build a LiveKit-based voice orchestration framework from scratch and migrate every existing agent onto it, at a scale of roughly 10 million calls a month. It's a rare combination of greenfield architecture and immediate production scale — the system you design starts carrying real call volume almost straight away. You'd report to the founder or CTO, with no layer in between.

What you'll do

Build the in-house voice orchestration framework on LiveKit from scratch
Migrate all existing AI agents off the current voice provider onto the new LiveKit-based system
Own prompt engineering and conversation design for voice agents handling millions of consumer interactions
Optimize real-time voice performance — latency, turn-taking, and state management — at production scale
Build testing and evaluation frameworks that continuously improve agent performance and business outcomes

What we're looking for

These are hard requirements. Check each one that applies to you.

Please note: every item above is a hard requirement for this role. Applications that don't meet all of them will be rejected. We recommend applying only if you check every box.

Don't check every box? No worries — submit your info here and we'll keep you in mind for roles that could be a better fit.

Submit Your Info

Nice to have

Experience at a voice AI or conversational AI company
Having run ASR/STT/TTS pipelines in production
Fintech, collections, or other heavily regulated consumer domains
Building evaluation harnesses for LLM or agent systems

Tech stack

Node.js and JavaScript/TypeScript
Python
LiveKit and real-time voice orchestration
ASR / STT / TTS and LLMs
AWS
SQL

Compensation & logistics

$200k–$300k base + 0.25–1% equity
Hybrid — 3 days a week in the Flatiron/NoMad office in New York City
Open to visa transfers (for example OPT or H-1B transfers); no new visa sponsorship
Reports to the founder or CTO
The process includes a paid take-home assignment of 3–5 hours, relevant to voice AI / LiveKit infrastructure
Full-time · hiring 1 for this role

AI Agent Engineer