Principal Machine Learning Tech Lead, Blue River Technology

• We are looking for an experienced, motivated, and collaborative technical leader to help build the future of autonomous agriculture
• 8-12 years of experience developing high-performance ML systems
• Experience shaping the technical vision for larger projects and providing guidance while contributing to core capabilities
• Strong leadership and communication skills, and demonstrated experience in leading and mentoring teams
• Hands-on work experience with GPU-based computing on large data sets
• Extensive experience with deep learning frameworks (e.g., PyTorch)
• Bachelor’s or Master’s Degree in Computer Science or a related field, graduate degree preferred

Desirable

• We are looking for a highly skilled and innovative Principal Machine Learning Tech Lead to join the team responsible for building the perception stack for Blue River Technology's autonomous tractor initiative
• The hire will be responsible for leading development on ML models, training and testing pipelines to enhance system safety and performance in challenging production environments
• This role will range from developing initial prototypes to scaling for production
• Lead the design, development, and deployment of machine learning and computer vision models that power real-time perception for autonomous agricultural robots
• Own the full ML lifecycle—from dataset design and data collection through training, evaluation, model optimization, and deployment on embedded platforms
• Collaborate cross-functionally with robotics engineers, software developers, and hardware teams to integrate perception systems into field-ready machines
• Lead a team of CVML engineers, fostering technical growth, strong engineering practices, and a culture of experimentation and delivery
• Define and execute the ML roadmap, balancing research exploration with production needs to meet performance, safety, and robustness targets in challenging environments
• Prototype and validate new approaches quickly, using both real-world and synthetic data to improve detection, segmentation, and scene understanding under variable conditions (lighting, occlusion, dust, etc)
• Continuously evaluate model performance in the field and refine approaches to maximize reliability and generalization across crop types, terrains, and