Senior Voice AI Engineer

Function

We're looking for a knowledgeable engineer to join our client's team, focusing on cutting-edge Voice AI technologies. This role involves collaborating with a diverse group of AI Dev Engineers and Data Scientists to develop a sophisticated voice bot that will directly engage with clients. Utilizing a hybrid speech stack featuring open-source speech models, the goal is to deliver optimal real-time performance.

Responsibilities and Role

Your contributions will be vital in enhancing the Voice Bot capabilities. You'll be instrumental in designing and implementing streaming pipelines for various audio tasks, and you'll take ownership of turn-taking functionalities which include managing interruptions and improving overall performance. Your role will also involve integrating telephony and app channels, ensuring system reliability through various technical optimizations.

• Design and implement streaming pipelines: audio ingestion, VAD/endpointing, STT, orchestration/LLM, streaming TTS.
• Manage turn-taking features, including barge-in, interruptions, endpoint tuning, and latency/accuracy trade-offs.
• Integrate telephony and application channels (SIP/WebRTC/CPaaS).
• Implement reliability features such as retries, backpressure, and rate limiting.

In addition, you will collaborate with Data Scientists to define and develop production-ready solutions, contributing to the automation of various aspects of the machine learning pipeline and supporting the IT production team in optimizing the target environment.

Profile of the Ideal Candidate

We are seeking a candidate with at least 4 years of relevant experience in engineering and a strong technical foundation. You will need to demonstrate proficiency in Python along with a systems programming language (preferably Go, Rust, or C++), as well as familiarity with streaming audio technologies and ASR/TTS stacks.

• 4+ years of engineering experience, with 2+ years of deploying streaming speech solutions in production.
• Strong knowledge of real-time STT/TTS, integration with telephony systems, and experience with various WebSockets or gRPC.
• Familiarity with open-source STT/TTS and vendor APIs, along with Docker/VM environment experience.
• Knowledge of CI/CD practices and experience with PostgreSQL.

Fluency in English is mandatory, and knowledge of Dutch or French is a plus. You should possess creativity, strong communication skills, and an eagerness for continuous learning and problem-solving.

Soft Skills and Business Experience

The ideal candidate should have a solid understanding of agile methodologies and demonstrate exceptional soft skills, including:

• Excellent oral and written communication abilities.
• A results-driven attitude with a keen focus on attention to detail.
• Innovative problem-solving capabilities and a proactive approach to continuous knowledge improvement.
• A constructive mindset that welcomes feedback and embraces change.

If you are ready to contribute to an engaging and dynamic environment, we encourage you to explore this opportunity with our client.

Contactperson & Reference

• Reference #: INW25530
• Casper Moyaert
• casper.moyaert@i4m.be