Applied ML Engineer San Francisco - Full time $140K - $240K - Offers Equity

$140,000 - $240,000/Yr

Speak Agency - San Francisco, CA

Full-time - Mid Level

San Francisco, CA

51-100 employees

Performing Arts, Spectator Sports, and Related Industries

About the position

At Speak, we are on a mission to revolutionize language learning by providing an AI-powered experience that allows users to practice speaking without needing a human partner. Our focus is on teaching English and Spanish to the next billion learners, addressing the challenges faced by those who struggle to find conversational partners. We have developed a unique approach that combines phoneme recognition technology with personalized learning experiences, enabling users to receive effective pronunciation feedback. As a Machine Learning Engineer, you will play a crucial role in this journey by taking ownership of the end-to-end modeling pipeline, from training and experimentation to deployment and monitoring. You will collaborate closely with product teams to design innovative learning experiences and measure the efficacy of our models in real-world applications. In this dynamic role, you will be responsible for training and deploying phoneme recognition models, expanding our Pronunciation Coach, and building an assessment system that provides nuanced feedback on various aspects of pronunciation. You will also maintain data infrastructure, including audio data pipelines and training datasets, while supporting the broader Speech & ML team in developing and localizing ASR models. This is an exciting opportunity to contribute to a small, agile team that is making a significant impact on language learning for millions of users worldwide. Join us at this pivotal moment as we continue to grow and expand our reach across global markets.

Responsibilities

• Training and deploying phoneme recognition models end-to-end, including monitoring, performance tracking, and retraining.
• Expanding our Pronunciation Coach to provide precise feedback and integrate more broadly across our learning platform.
• Tracking metrics to measure performance of existing phoneme models across markets.
• Building out an assessment system to provide nuanced feedback on pronunciation, intonation, prosody, and more.
• Building and maintaining data infrastructure; i.e. audio data pipelines, training/evaluation datasets creation and management, labeling/active learning loop.
• Supporting the broader Speech & ML team (e.g. developing and localizing ASR models).

Requirements

• Extensive experience training and deploying custom deep learning models to production (experience with audio/speech strongly preferred).
• Proficiency in Python and common Deep Learning frameworks like PyTorch.
• Strong communication skills and the ability to explain complex ML concepts to non-technical stakeholders.
• Sharp product sense and an ability to think broadly and cross-functionally about model quality in the context of user experience.

Benefits

• Opportunity to work with a fantastic, tight-knit team.
• Significant impact on the company's direction and growth.
• Global exposure with opportunities to interact with users in various countries.
• Supportive work environment focused on personal and professional growth.