ML Research Engineer, Speech

Echo Neurotechnologies • San Francisco, CA

15d

About The Position

Echo Neurotechnologies is an exciting new startup in the Brain-Computer Interface (BCI) space, driving innovation through advanced hardware engineering and AI solutions. Our mission is to deliver cutting-edge technologies that restore autonomy to people living with disabilities and improve their quality of life. Join a small, dedicated team of knowledgeable and motivated professionals. Our early-stage environment offers the opportunity to take ownership of broad decisions with significant and long-lasting impact. We emphasize continuous learning and growth, fostering cross-functional collaboration where your contributions are vital to our success. We are seeking an experienced ML Research Engineer, with a focus on speech modeling, to join our team. The person who fills this role will apply ML/AI principles and practices to designing and developing flexible models of speech for use within brain-computer interface applications to have real-world impact on patients with severe disabilities. They will need to have extensive experience with speech modeling & feature engineering, time-series analysis, and building custom ML models from scratch.

Requirements

• Bachelor's or Master's degree in Math, Engineering, Neuroscience, or other relevant field
• A minimum of 2 years of academic or industry research experience working with speech datasets
• Experience performing feature engineering with speech signals
• Proficiency in Python and PyTorch
• Excellent communication and collaboration skills to work effectively on cross-functional and interdisciplinary teams
• Experience having technical ownership over at least one successfully implemented collaborative project

Nice To Haves

• A PhD in Math, Engineering, Neuroscience, or other relevant field
• Experience developing speech synthesizers
• Experience working with neural signals, especially for brain-computer interface applications

Responsibilities

• Design and assess varying model architectures and modeling strategies to maximize the accuracy and robustness of models for decoding speech from brain activity
• Explore the use of state-of-the-art speech features and representations within neural-decoding pipelines, informed by the fields of speech science and functional neurophysiology
• Develop pipelines for synthesizing personalizable and naturalistic speech from text and from brain activity
• Build algorithms to analyze intact and impaired speech signals to detect biomarkers related to underlying diseases and disabilities
• Work collaboratively within a small team to develop models, design R&D workflows, and translate scientific findings into product implementations
• Contribute to best practices for reliability, observability, reproducibility, and scientific rigor across the R&D ecosystem
• Maintain versioned, clear, and highly documented code, analysis pipelines, and results for maximum interpretability and reproducibility

Benefits

• An opportunity to work on exciting, cutting-edge projects to transform patients’ lives in a highly collaborative work environment.
• Competitive compensation, including stock options.
• Comprehensive benefits package.
• 401(k) program with matching contributions.

What This Job Offers

• Job Type: Full-time
• Career Level: Mid Level