Research Engineer – Robotics & Embodied AI

Elorian • Palo Alto, CA

8d

About The Position

Elorian is a well-funded, early-stage AI lab building frontier multimodal AI models. Founded by former DeepMind researchers, the company has a team of 20 scientists and engineers and raised $55 million in seed funding.

The Research Engineer will apply robotics expertise to foundation-scale AI. Working closely with the core foundation model research team, this person will bring physical-world priors into multimodal models, address real-world robotics challenges, and develop foundational capabilities that generalize across robotics applications. The role focuses on visual reasoning, spatial understanding, robotics data, and embodied reasoning rather than isolated robot hardware or narrow policies.

Requirements

• Strong background in robotics research or engineering, with firsthand experience in robotic perception, planning, control, or robot learning.
• Genuine interest in applying robotics expertise to large-scale deep learning, multimodal architectures, and foundation model research.
• Solid machine learning experience and hands-on familiarity with modern LLMs, multimodal foundation models (VLMs), or transformer architectures.
• Strong research and problem-solving skills, including the ability to independently investigate open-ended, complex technical challenges.

Nice To Haves

• Experience with robotics simulation environments such as Isaac Sim or MuJoCo for large-scale data generation, policy evaluation, or model pre-training.
• Experience with robotics datasets, trajectory annotation, task decomposition, and benchmark evaluation for physical systems.
• Experience applying, fine-tuning, or developing LLMs, VLMs, or Vision-Language-Action models for robotics problems.
• Background working with real-world robotic systems in manipulation, navigation, or autonomous agents, and understanding the practical limits of deploying policies to physical hardware.
• Deep expertise in robotic planning, 3D spatial reasoning, affordance modeling, or embodied intelligence.

Responsibilities

• Advance and evaluate visual reasoning, 3D spatial understanding, and physical affordance capabilities within multimodal foundation models for downstream robotic perception and control.
• Design scalable robotics data pipelines, including data collection, teleoperation, trajectory annotation, automated quality assessment, and evaluation frameworks for multimodal robotic data.
• Investigate and implement approaches to robotic task planning, hierarchical reasoning, and decision-making using multimodal foundation models and Vision-Language-Action architectures.
• Collaborate with foundation model researchers and robotics partners to identify real-world bottlenecks, evaluate model capabilities, and develop foundational AI solutions for the robotics ecosystem.

What This Job Offers

• Job Type: Full-time
• Education Level: No Education Listed