Research Engineer, Robotics & Physical AI

About the Opportunity

Invyte.ai is partnering with an early-stage Physical AI company building intelligent systems that learn from video, multimodal data, and real-world robotic interactions. We’re looking for talented Research Engineers and Research Interns in Robotics & Multimodal AI who are passionate about advancing the intersection of AI, robotics, and real-world intelligence. This is a hands-on, research-driven opportunity to work on cutting-edge problems in robot learning, computer vision, vision-language models (VLMs), video understanding, and embodied AI. You’ll go beyond reading papers and building prototypes—you’ll design experiments, reproduce promising research, develop rigorous evaluations, and turn successful methods into systems that work in the real world. We value demonstrated technical ability, research depth, and what you’ve built over titles or years of experience. Exceptional recent graduates, PhD researchers, interns, and experienced engineers are all encouraged to apply.

Key Responsibilities

• Research and develop models for robotics, robot learning, multimodal AI, and computer vision.
• Work extensively with video, visual observations, and real-world multimodal robotic data.
• Explore vision models, vision-language models (VLMs), imitation learning, and emerging Physical AI approaches.
• Read recent research papers, reproduce state-of-the-art methods, and design novel experiments.
• Build robust evaluations, benchmarks, and metrics to measure model and system performance.
• Use feedback and data from deployed systems to identify failures and continuously improve models.
• Translate research breakthroughs into practical solutions for real-world robotic applications.

Requirements & Tech Stack

• Strong demonstrated work in one or more of the following areas:
• Robotics, Robot Learning, or Embodied AI
• Physical AI and Imitation Learning
• Multimodal Learning and Vision-Language Models
• Computer Vision and Video Understanding
• Model Post-training, Evaluation, and Benchmark Design
• Strong Python and PyTorch skills.
• Experience designing experiments, implementing research ideas, or reproducing research papers.
• A strong problem-solving mindset and the ability to work on open-ended research challenges.
• A genuine interest in building AI systems that interact with and learn from the physical world.

Good to have:

• ROS (Robot Operating System)
• Hands-on robotic systems experience
• Experience deploying models in real-world environments

Qualifications

We welcome exceptional recent graduates, PhD researchers, research interns, and engineers with strong projects, publications, open-source contributions, or demonstrated technical achievements. Prior industry experience is not mandatory.