Founding Research Engineer

Mbodi; New York (NY); On-site; USD 180,000 - 280,000; Full time

Job summary

Mbodi AI, an embodied AI startup, seeks a Founding Research Engineer to build the intelligence behind robots that can learn and adapt to new tasks. You will push the boundaries of agentic AI, enabling robots to understand language, reason about the physical world, and execute complex tasks reliably on real robots.

You’ll explore state-of-the-art foundation models, craft novel algorithms, and deploy research in a hands-on engineering role.

Qualifications

• Strong AI/ML foundations including transformers and foundation models.
• Ability to develop and evaluate novel AI algorithms for robots.
• Proven Python and ML framework experience (e.g., PyTorch).

Responsibilities

• Develop agent architectures for task planning, decision-making, and autonomous error recovery.
• Research and implement methods for robots to learn from language, demonstrations, and interaction.
• Advance multimodal perception and foundation model techniques for physical-world tasks.
• Translate research into working robotic systems and validate in real hardware.

Skills

Agentic AI

Multimodal models

Vision-language models

Robot learning

Reinforcement learning

Python

PyTorch

Systems thinking

Startup mindset

Education

Advanced degree in CS/ML/Robotics

Tools

PyTorch

TensorFlow

Job description

About Mbodi

Join Mbodi AI (YC X25), an AI robotics startup founded by two former Googlers committed to pushing the boundaries of physical intelligence. Mbodi is an embodied AI platform that makes robots learn and operate like humans. Anyone can teach robots new skills by talking to them — and execute those skills reliably in production within minutes.

We are pioneering the next wave of robotics, where AI agents meet physical-world applications. Backed by top investors and working with global industrial partners such as ABB, we’re redefining what’s possible in robotics and automation.

Role Overview

We’re looking for a Founding Research Engineer to join our core team and develop the intelligence behind robots that can learn and adapt to new tasks.

You’ll work at the frontier of agentic AI and embodied intelligence, developing systems that enable robots to understand natural language, reason about the physical world, learn from demonstrations, and execute complex tasks reliably.

You'll explore and integrate state-of-the-art foundation models while developing novel algorithms and architectures to push the boundaries of robot autonomy.

This is a research-driven, hands-on engineering role where you'll take ideas from early experimentation to real-world deployment on physical robots. As a founding team member, you'll help shape our core technology, research direction, and engineering culture from the ground up.

What You’ll Do

• Agentic AI & Reasoning: Develop intelligent agent architectures for task planning, decision-making, skill composition, and autonomous error recovery.
• Embodied AI & Robot Learning: Research and implement approaches that enable robots to learn new skills from natural language, demonstrations, and interaction.
• Multimodal Intelligence: Leverage and advance vision-language models, foundation models, and perception systems for physical-world understanding and action.
• Applied Research: Explore emerging techniques in generative AI, robotic foundation models, and learning-based control, translating promising research into working systems.
• Real-World Validation: Design experiments, evaluate model performance, and deploy algorithms on physical robots to improve generalization, reliability, and autonomy.
• Technical Leadership: Shape core AI architecture, research directions, and technical strategy as an early member of the founding team.

What We’re Looking For

• Strong AI/ML Foundations: Deep understanding of modern machine learning, including transformers, foundation models, and generative AI architectures.
• Research & Implementation: Proven ability to develop, implement, and evaluate novel AI algorithms, with strong Python and ML framework experience (e.g., PyTorch).
• Relevant Experience: Hands-on experience in one or more of agentic AI, multimodal models, vision-language models, robot learning, reinforcement learning, or related areas.
• Systems Thinking: Ability to build end-to-end AI systems, connecting models, reasoning, perception, and real-world execution.
• Research Mindset: Strong experimental intuition, ability to identify promising research directions, and willingness to challenge existing approaches.
• Startup DNA: High agency, intellectual curiosity, and a strong bias toward building and shipping in a fast-paced environment.

Preferred Qualifications

• Experience with embodied AI, robotics, imitation learning, diffusion policies, or vision-language-action models.
• Experience building agentic systems involving LLMs/VLMs, tool use, planning, or multi-agent orchestration.
• Experience training, fine-tuning, or adapting foundation models for specialized applications.
• Experience deploying ML systems on physical robots or other real-world, latency-sensitive systems.
• Advanced degree in Computer Science, Machine Learning, Robotics, or a related field, or equivalent research and engineering experience.