Voice AI Architect — Real-Time Production Leader

Neurons Lab; Warszawa; Hybride; PLN 260 000 - 480 000; Plein temps

Résumé du poste

Neurons Lab in Poland is seeking an experienced AI architect to own the technical architecture and delivery of a real-time voice copilot for veterinary care. You will drive PoC to production, optimize latency and cost, and ensure strict data isolation while keeping stakeholders aligned with the SOW.

The role leads the AI Engineer and the pod, conducting knowledge transfer, **Langfuse** tracing, and end-to-end epics with a production-ready mindset.

Qualifications

• Experience shipping real-time voice or speech products in production.
• Strong track record in latency/reliability improvements.
• Ability to lead a small AI engineering team and coordinate with clients.

Responsabilités

• Own the full pipeline: streaming STT, LLM field extraction, and deployment.
• Reduce latency (P95) from ~6s toward ~2s and fix post-processing delays.
• Run model A/B tests with golden-set evaluation for accuracy.
• Own cost evaluation and optimization plans; monitor production metrics.

Connaissances

Real-time voice pipelines

LLM engineering

Observability & evals

AWS & Bedrock

Python full-stack

English communication

Formation

Advanced CS/ML degree

Outils

Langfuse

AWS SES

Chrome-extension integration

Description du poste

Neurons Lab in Poland is seeking an experienced AI architect to own the technical architecture and delivery of a real-time voice copilot for veterinary care. You will drive PoC to production, optimize latency and cost, and ensure strict data isolation while keeping stakeholders aligned with the SOW.

The role leads the AI Engineer and the pod, conducting knowledge transfer, **Langfuse** tracing, and end-to-end epics with a production-ready mindset.