Senior AI Engineer

Voice AI/NLP Expert

We are seeking a Voice AI/NLP expert to lead the design and deployment of a real-time, on-premises conversational AI pipeline for call automation. You will be responsible for architecting and implementing streaming Speech-to-Text (ASR), local LLM/NLU, and Text-to-Speech (TTS) solutions using open-source models such as Whisper, Vosk, Coqui, and Llama. The ideal candidate has deep hands-on experience with real-time audio processing, low-latency inference, and building scalable microservice-based AI systems. Strong Python skills and proven expertise in integrating AI pipelines with telephony/media servers (such as LiveKit, SIP, or WebRTC) are essential, as is a strong commitment to privacy and data security—our entire stack runs offline and on-premises.

You will work closely with backend and DevOps engineers to deliver a robust, horizontally scalable, multi-tenant voice platform that handles high call volumes and sensitive data. Experience in AI observability, performance tuning, and continuous evaluation of speech/NLP models is key. Familiarity with tool-calling, call routing, and integrating AI agents into contact center workflows is highly valued. If you have a track record of deploying and optimizing conversational AI in production environments, especially for privacy-first or enterprise solutions, we want to hear from you.

Requirements

• Expertise in Voice AI/NLP with proven experience designing and deploying real-time conversational AI pipelines.

• Strong proficiency in Python and building scalable, microservice-based AI systems.

• Hands-on experience with streaming Speech-to-Text (ASR), LLM/NLU, and Text-to-Speech (TTS) using open-source models (e.g., Whisper, Vosk, Coqui, Llama).

• Deep knowledge of real-time audio processing and low-latency inference.

• Proven experience integrating AI pipelines with telephony/media servers (e.g., LiveKit, SIP, WebRTC).

• Strong understanding of privacy-first, on-premises deployments and data security best practices.

• Experience collaborating with backend and DevOps engineers to deliver scalable, multi-tenant voice platforms.

• Knowledge of AI observability, performance tuning, and continuous model evaluation.

• Familiarity with tool-calling, call routing, and integration of AI agents into contact center workflows.

• Demonstrated track record of deploying and optimizing conversational AI in production, ideally for enterprise or privacy-first environments.