Manager of Multi-Modal Language Action Models

Join Zoox, a cutting-edge autonomous driving company, as the Manager of Multi-Modal Language Action Models. In this role, you will lead a team of talented Machine Learning Engineers, oversee the full lifecycle of MLLM development, and drive the deployment of complex MLLM solutions to internal customers. You will also collaborate with cross-functional teams, create development opportunities for your team members, and establish team goals and priorities. This position offers a comprehensive benefits package, including paid parental leave, health insurance, and work-from-home opportunities.

• Lead and manage a team of talented Machine Learning Engineers specializing in Multi-Modal Large Language Models (MLLMs) for concrete applications.
• Oversee the full lifecycle of MLLM development, with a specific focus on post-training methodologies to optimize models for autonomy tasks.
• Drive the deployment of complex MLLM solutions to internal customers, specifically solving tangible offline autonomy problems within the Zoox ecosystem.
• You do not need to match every listed expectation to apply for this position. Here at Zoox, we know that diverse perspectives foster the innovation we need to be successful, and we are committed to building a team that encompasses a variety of backgrounds, experiences, and skills
• Proven track record of deploying ML/MLLM models to production or internal customers to solve complex, real-world problems, ideally within the autonomy or robotics domain
• 3+ years of experience leading and managing engineering or research teams focused on AI/ML
• 5+ years of professional experience in Machine Learning or Deep Learning, with a focus on large-scale model applications
• Strong organizational skills to manage complex project schedules, staffing needs, and priorities effectively
• A passion for mentoring and developing high-performing team members
• Direct experience leading teams training Multi-Modal Large Language Models (MLLMs), with deep expertise in post-training techniques (e.g., instruction tuning, alignment, domain adaptation)