AI Research Scientist

RoboForce is an AI robotics company developing Physical AI–powered Robo-Labor for dull, dirty, and dangerous work. The company's robots are engineered for demanding industrial environments, with a focus on real-world deployment and scalability.

We are looking for a Senior / Staff AI Research Scientist, Foundation Models to advance robotic embodied intelligence. In this role, you will develop algorithms that enable robots to understand their environment, interpret and execute tasks, and communicate seamlessly with humans — with a particular focus on building and training world models that allow robots to predict, plan, and generalize across complex physical tasks.

• Design and deploy vision-language(-action) models (VLM/VLA) for contextual understanding and generalized robot action policies.

• Develop and train world models for action-conditioned prediction, long-horizon planning, and environment simulation — enabling robots to reason about the consequences of their actions before execution.

• Research approaches to improve world model fidelity using multi-modal inputs including vision, language, proprioception, and spatial representations.

• Develop foundation models with spatial reasoning capabilities to achieve high-precision robotic actions.

• Integrate multi-modal data sources (vision, language, speech, etc.) to enable natural human-robot communication.

• Optimize and deploy models as production-grade solutions on RoboForce robotic platforms.

• PhD degree in Machine Learning, Robotics, or related field, or Master's degree with 4+ years of relevant experience.

• Proficiency in Python and deep learning frameworks (e.g., PyTorch, JAX).

• Expertise in large foundation models (VLM, VLA, etc.).

• Strong understanding of world model architectures and action-conditioned generative modeling for robot learning.

• Decent understanding of multimodal models, modern ML architectures (transformers, diffusion models, etc.).

• Requires 5 days/week in-office collaboration with the teams.

• Experience with video generation or prediction models (e.g., diffusion-based video models, autoregressive video transformers) and their application to world modeling or synthetic data generation for robot learning.

• Strong publication record at top conferences (NeurIPS, ICML, CVPR, ICCV, CoRL, ICRA, or equivalent).

• Expertise in neural network deployment (e.g., TensorRT) and GPU programming with CUDA.

• Proven ability to design scalable experimentation and data pipelines.