Research Scientist, Dexterous Manipulation & Robot Learning

Lila Sciences•Cambridge, MA

2d•$176,000 - $304,000

About The Position

As a Robotics Scientist at Lila, you will lead the research and development of autonomous robotic systems that serve as the intelligent physical infrastructure of our scientific superintelligence platform. You’ll develop novel algorithms and deploy intelligent robotic solutions that interact seamlessly with human scientists and complex lab environments. Your work will accelerate our mission by enabling fully autonomous workflows for scientific discovery, combining cutting-edge robotics, machine learning, and systems engineering. Lila Sciences is the world’s first scientific superintelligence platform and autonomous lab for life, chemistry, and materials science.  We are pioneering a new age of boundless discovery by building the capabilities to apply AI to every aspect of the scientific method.  We are introducing scientific superintelligence to solve humankind's greatest challenges, enabling scientists to bring forth solutions in human health, climate, and sustainability at a pace and scale never experienced before. Learn more about this mission at www.lila.ai If this sounds like an environment you’d love to work in, even if you only have some of the experience listed below, we encourage you to apply.

Requirements

Ph.D. in Robotics, Machine Learning, Computer Science, or a related field with demonstrated expertise in foundation models for robotic learning
Advanced proficiency in reinforcement learning, diffusion-based methods, imitation learning, and adaptive learning algorithms for robotic manipulation
Expert-level experience with machine learning frameworks (PyTorch, TensorFlow) and deep learning architectures for developing foundation models, with specific expertise in diffusion-based generative models for robotics
Proven track record of developing multi-modal perception systems integrating tactile, visual, language and other contextual sensing for intelligent robotic agents
Strong publication record in robot learning, demonstrating innovative approaches to trust calibration, contextual learning, and generative robotic skill learning

Nice To Haves

Research contributions to foundation models and diffusion methods in robotics
Experience with large-scale machine learning model development, particularly generative and diffusion-based approaches
Expertise in human-in-the-loop learning, correction-based training paradigms, and diffusion-guided skill transfer
Demonstrated ability to translate theoretical machine learning research, especially diffusion and generative models, into practical robotic implementations

Responsibilities

Pioneering approaches for precise and dexterous robotic manipulation that leverage foundation models, reinforcement learning, diffusion-based methods, and human guidance to enable adaptive and intelligent robotic systems capable of complex tasks across diverse scientific environments
Developing novel human-robot interaction frameworks that incorporate imitation learning, and learning from human guidance, feedback, demonstrations and corrections, creating intelligent robotic agents that can seamlessly integrate with human scientific workflows and rapidly adapt to new experimental contexts
Advancing dexterous manipulation research through cutting-edge machine learning approaches, including diffusion models and adaptive learning algorithms, that synthesize multi-modal sensing (tactile, visual, and language) to develop generative skill representation sand sophisticated motor learning policies for intelligent robotic systems
Designing autonomous robotic systems with trust calibration mechanisms, enabling intelligent agents that can dynamically adjust their behaviors based on contextual information in complex scientific tasks