Lead AI Research Scientist (Robotics)

Accountabilities:

• Lead the research, development, and adaptation of Vision-Language-Action (VLA) models for advanced robotic applications.
• Fine-tune open-weight VLA models, including OpenVLA, Octo, and π0, for robotic navigation and manipulation tasks.
• Develop multimodal representations connecting camera observations, natural-language instructions, spatial reasoning, and robotic actions.
• Design action-tokenization strategies and spatial waypoint generation methods to improve robotic behavior and decision-making.
• Build and optimize parameter-efficient fine-tuning pipelines using LoRA and other PEFT techniques.
• Leverage Vision-Language Models (VLMs), Vision Transformers (ViT), and large multimodal models to enhance robotic perception and intelligence.
• Integrate AI-driven visual perception and reasoning capabilities into robotic decision-making systems.
• Explore cross-embodiment learning approaches and datasets such as Open X-Embodiment to improve generalization across different robotic platforms.
• Collaborate with robotics control and embedded systems teams to translate AI model outputs into executable robotic behaviors.
• Lead research planning, technical evaluations, experimentation, and continuous model performance improvements.
• Mentor and guide researchers and engineers throughout model adaptation, training, evaluation, and deployment processes.
• Translate research findings into practical, scalable AI solutions for real-world robotic applications.

Requirements:

• Master's or Ph.D. degree in Computer Science, Artificial Intelligence, Robotics, Machine Learning, or a related technical discipline.
• At least 4 years of relevant professional experience in multimodal AI, foundation models, vision-language models, imitation learning, or robotic learning.
• Strong hands-on experience developing, training, and fine-tuning large multimodal models, Vision Transformers, or related deep learning architectures.
• Advanced proficiency in Python and deep learning frameworks such as PyTorch or JAX.
• Practical experience working with Vision-Language Models (VLMs) or Vision-Language-Action (VLA) architectures.
• Familiarity with open-weight VLA models such as OpenVLA, Octo, and π0.
• Experience using Hugging Face tools and model training, adaptation, and fine-tuning workflows.
• Solid understanding of parameter-efficient fine-tuning techniques, including LoRA and related approaches.
• Strong knowledge of multimodal representation learning, computer vision, visual perception, and language-conditioned action generation.
• Familiarity with robotic imitation learning, policy learning, and/or reinforcement learning.
• Understanding of cross-embodiment robotic learning and datasets such as Open X-Embodiment is considered an advantage.
• Demonstrated ability to lead technical research initiatives, assess experimental outcomes, and translate findings into practical AI capabilities.
• Strong analytical, problem-solving, and communication skills, with the ability to collaborate effectively across multidisciplinary research and engineering teams.
• Proven ability to provide technical direction, mentor colleagues, and drive research projects from experimentation to implementation.

Benefits:

• Fully remote working arrangement through a US-based research team, subject to approved employer work locations.
• Opportunity to lead the development of emerging Physical AI and Vision-Language-Action technologies.
• Exposure to cutting-edge research at the intersection of multimodal AI, foundation models, and robotics.
• Opportunity to explore innovative approaches to language-conditioned robotic navigation and manipulation.
• Collaboration with AI researchers and robotics engineers on advanced, real-world applications.
• Significant influence over technical direction, research priorities, and the evolution of intelligent robotic systems.
• Contract-based employment arrangement.
• Alternative hybrid working arrangement may be available in New Delhi, India, subject to employer approval.