Senior/Staff AI Research Engineer

Hume AI is seeking talented researchers and engineers interested in working with our AI research team to build state-of-the-art speech-language models (SLMs). Our new SLM training method—reinforcement learning from human expression (RLHE)—learns human preferences from behavior in millions of audio and video recordings, making SLMs superhumanly helpful, interesting, funny, eloquent, honest, and altruistic. Join us in the heart of New York City and contribute to our endeavor to ensure that AI is guided by human values, the most pivotal challenge (and opportunity) of the 21st century.

About Us**

Hume AI is a Series B startup dedicated to building artificial intelligence that is directly optimized for human well-being. As the first company to release speech language models, we’re focused on expanding our research to encompass audio understanding models and evaluation platforms for enterprises.

Our goal is to enable a future in which technology draws on an understanding of human emotional expression to better serve human goals. As part of our mission, we also conduct groundbreaking scientific research, publish in leading scientific journals like Nature, and support a non-profit, The Hume Initiative, that has released the first concrete ethical guidelines for empathic AI (). You can learn more about us on our website () and read about us in WIRED, Forbes, and Venturebeat.

About the Role**

We are looking for talented researchers and engineers interested in working with our team to build state-of-the-art audio models and evaluation tools. In this role, you will build out software systems that support distributed model training, inference, and benchmarking as well as massive-scale data collection, storage, preprocessing, and analysis. You will be working to solve some of today’s most exciting AI research problems at industry scale.

Requirements

-
Expertise in the Python ecosystem and popular ML libraries and tools (e.g. PyTorch)

-
Experience writing robust and maintainable production-ready code

-
Comfort iterating quickly on new and uncertain research directions

• 2+ years of experience training and/or fine-tuning transformer models with large-scale datasets of text, audio, image, and/or video data

Application Note

Please apply only to the position that best aligns with your qualifications. If you submit multiple applications or have applied within the past 6 months, only your initial submission will be considered.

Annual Salary

$180,000—$350,000 USD