Senior Machine Learning Engineer, Applied AI Modeling

About the role

Mozilla is integrating advanced AI and Large Language Model capabilities into the Firefox browser to enhance user experiences while prioritizing privacy. This role involves conceptualizing, building, and deploying these new features. The successful candidate will own the design and execution of machine learning workflows that power intelligent browser functionalities. You will be responsible for translating high-level product goals into robust, production-grade AI solutions. A core part of this position is ensuring that models operate efficiently and securely within the constraints of a modern web browser. You will work at the intersection of user privacy and cutting-edge generative technology. This role demands a proactive approach to solving complex engineering challenges in a distributed team environment.

Key facts

What you'll do

• Architect and implement large language models to facilitate secure and responsive interactions directly within the Firefox browser interface.
• Engineer and integrate retrieval-augmented generation, advanced summarization, and intelligent classification methods to optimize browser workflows.
• Execute the complete machine learning lifecycle, utilizing specialized tools for experimentation, rigorous evaluation, and scalable deployment of models.
• Collaborate closely with product managers and software engineers to define, build, and launch user-facing AI features that align with Mozilla's core mission and privacy principles.
• Design and analyze controlled experiments to quantitatively measure model accuracy, latency, and user satisfaction in real-world browser environments.
• Contribute to technical documentation and actively participate in collaborative code reviews to ensure high standards of quality and knowledge sharing.
• Evaluate and select appropriate data sources and preprocessing techniques to build high-quality datasets for training and fine-tuning language models.
• Optimize model performance and resource usage to meet the strict performance and efficiency requirements of a global-scale web browser.
• Implement monitoring strategies to track model behavior and data drift over time, ensuring sustained reliability and user trust.
• Mentor junior engineers and researchers by providing technical guidance and fostering best practices in machine learning development.

Requirements

• Hold a minimum of 4 years of professional experience in applied machine learning, with a demonstrated focus on natural language processing or generative AI technologies.
• Possess practical, hands-on experience in fine-tuning and rigorously assessing large language models using contemporary open-source libraries and tooling.
• Demonstrate a solid conceptual and practical understanding of prompt engineering techniques, embedding-based retrieval systems, and robust evaluation methodologies for generative AI applications.
• Show a proven track record of managing the full lifecycle of model development, from initial data analysis and feature engineering to the delivery of production-ready model outputs.
• Have a history of successfully creating and deploying user-facing AI features where non-negotiable requirements include privacy preservation, low latency, and high usability.
• Exhibit effective communication and collaboration skills necessary for working effectively with cross-functional teams in a fully distributed work setting.
• Bring a strong foundation in machine learning fundamentals, including model training, validation, and inference optimization.
• Show commitment to writing clean, maintainable, and well-documented code that adheres to engineering best practices.

Nice to have

• Demonstrate familiarity with on-device model optimization strategies and privacy-preserving machine learning techniques to further enhance user control over data.
• Bring prior experience working within a browser environment or with web technologies, enabling faster integration and more effective troubleshooting.
• Show a history of meaningful contributions to open-source machine learning frameworks, libraries, or public model repositories that benefit the broader community.

Practical notes

The position is based in the United States and requires eligibility to work in the country without sponsorship. The standard work schedule is 40 hours per week. Compensation is structured into distinct tiers based on the employee's location within the United States, with specific ranges published for US Tier 1, Tier 2, and Tier 3 locations. These ranges reflect regional variations in cost of living and market standards. Candidates are encouraged to apply if they meet the core requirements and are passionate about Mozilla's vision for an open internet powered by responsible AI. The offered benefits package is comprehensive and includes performance-based bonuses, extensive medical, dental, and vision insurance, generous retirement contribution plans, designated wellness days, flexible paid time off, a home office stipend, a dedicated professional development budget, and structured parental leave policies.