Senior Machine Learning Engineer (Voice & Video AI)

The role

You will build conversational voice and video AI agents that can see, listen, speak, and respond in real time. You will take models from evaluation and fine-tuning into production on GPU hardware and own features across the full stack.

What you will do

• Evaluate, adapt, and deploy speech-to-speech and full-duplex video models.
• Improve natural interaction through turn-taking, interruption handling, and responses to cues.
• Track emerging research and benchmark promising approaches.
• Fine-tune models for languages, accents, and specialized domains.
• Build data and evaluation pipelines.
• Train and deploy models on GPU hardware across global infrastructure.
• Optimize streaming inference, GPU utilization, latency, and serving cost.
• Own features from model experiments through backend services, APIs, and customer portal interfaces.
• Use coding agents and harnesses to implement, test, debug, and review changes.

Required qualifications

• 6+ years in machine learning or related engineering.
• Production experience and depth in speech generation, speech-to-speech systems, video generation, or multimodal models.
• Strong Python and PyTorch skills.
• Hands-on experience adapting pretrained models.
• Experience with streaming or latency-sensitive systems.
• Experience with rigorous evaluation and tradeoffs between model quality, responsiveness, reliability, and cost.
• Comfort working beyond the ML codebase including backend and frontend changes.
• An agent-driven development mindset.

About the company

Telnyx is an industry leader building the future of global connectivity through a private, global, multi-cloud IP network and intuitive APIs. The company is financially stable and profitable.