Senior Machine Learning Engineer for Real-Time Voice AI
Design, train, and deploy production models for real-time voice AI systems. Work with cutting-edge language and audio models. Thrive in a fast-paced startup environment.
Key Highlights
Technical Skills Required
Benefits & Perks
Job Description
Senior ML Engineer, Real-Time Voice AI
San Francisco, Menlo Park (5 days per week)
$200,000 - $280,000 + meaningful equity
Visa sponsorship available
About the Company
Our client is a rapidly growing startup building real-time voice AI systems that need to perform under the most demanding conditions. They use cutting-edge language and audio models to deliver natural, intelligent interactions at scale. Backed by top investors, they have grown revenue quickly with a small team and are now expanding their core ML group to push the limits of real-time AI performance.
About the Role
This is a hands-on, high-ownership role for an ML engineer who wants to design, train, and deploy production models that actually ship and handle real-world constraints. You will work across LLMs, audio models, and adaptive reasoning components to support real-time voice interactions.
You will thrive here if you enjoy fast iteration, tight feedback loops, and building models that need to be accurate, robust, and incredibly fast. The team moves with genuine startup intensity, so this suits someone who enjoys long hours, deep work, and the thrill of pushing models to production quickly.
Key Responsibilities
- Train and fine-tune language and audio models for speed, accuracy, and real-time performance
- Build evaluation pipelines, benchmark datasets, and rigorous testing frameworks
- Deploy models end to end, from training pipelines to live monitoring and optimisation
- Work with engineering to integrate models into latency-sensitive production systems
- Run fast experimentation cycles to improve model behaviour and stability
- Contribute to research direction and long-term ML strategy
You Might Thrive If You
- Have trained and shipped production ML models, ideally in language or speech
- Are fluent in Python, PyTorch, and modern ML tooling
- Enjoy solving high-complexity problems under real-world constraints
- Like taking full ownership, from data to deployment
- Are comfortable in a high-intensity environment where the pace is fast and the hours are long
- Speak Mandarin fluently and enjoy working in a multilingual environment
- Want to help shape the future of real-time voice AI