Production AI Architect | Real-Time Voice Copilot Lead

Neurons Lab; Roma; In loco; EUR 90.000 - 130.000; Tempo pieno

Descrizione del lavoro

Neurons Lab in Rome, Italy, is seeking a senior AI engineer to own the architecture and delivery of the voice copilot product for the veterinary care client. You will manage real-time streaming STT, LLM field extraction, Chrome-extension delivery, and the AWS Bedrock infrastructure to production readiness.

The role focuses on reducing latency, validating accuracy, and controlling per-call costs, with ongoing knowledge transfer to the client team and Neurons Lab engineers.

Competenze

• Real-time voice AI in production — shipped at least one real-time voice or speech product to real users.
• 6+ years hands-on AI/ML engineering with strong recent LLM production practice.
• Latency improvements and concurrency fixes demonstrated on a live system.
• Consulting / client-facing seniority; calm and precise under detailed UAT scrutiny.

Mansioni

• Own the full pipeline: streaming STT, LLM field extraction, Chrome-extension delivery, and AWS infrastructure.
• Drive latency work: reduce P95 from ~6s toward ~2s and remove post-processing lag.
• Run model A/B tests with golden-set evaluation for name and email accuracy.
• Own evaluation and cost: Langfuse traces, dashboards, and per-call cost optimization.
• Harden for production: 5–10+ concurrent calls, data isolation, monitoring, and safe rollback.
• Ship epics end to end (SES email briefing); maintain a demo fallback for reliability.

Conoscenze

Real-time voice pipelines

LLM engineering

Observability

Low-latency inference

Python

English communication

Cost optimization

Strumenti

Langfuse

AWS Bedrock

SES

Chrome Extension

Descrizione del lavoro

Neurons Lab in Rome, Italy, is seeking a senior AI engineer to own the architecture and delivery of the voice copilot product for the veterinary care client. You will manage real-time streaming STT, LLM field extraction, Chrome-extension delivery, and the AWS Bedrock infrastructure to production readiness.

The role focuses on reducing latency, validating accuracy, and controlling per-call costs, with ongoing knowledge transfer to the client team and Neurons Lab engineers.