Research Scientist, Embodied AI & Robotic Systems

Ellenco Estágios e Treinamentos; Berkeley (CA); On-site; USD 150,000 - 210,000; Full time

Job summary

Ellenco Estágios e Treinamentos in Berkeley, CA seeks a Research Scientist in Embodied AI to architect robotic foundation models and lead the design of large‑scale learning frameworks for industrial manipulation and autonomous planning. You will build end‑to‑end training pipelines for multi‑modal robotics data, bridge the sim‑to‑real gap, and develop scalable inference loops for real‑time control.

Ideal candidates have Ph.D./M.Sc.

Qualifications

• Ph.D. or MSc from an accredited university.
• Proven track record at top robotics/AI venues (IROS/ICRA/RSS/CoRL).
• Fluency in C++ and Python at professional level.
• Experience deploying ML models on robotic systems (training, eval, integration).
• Deep knowledge of transformers, VLAs/VLMs, and diffusion models.

Responsibilities

• Architect robotic foundation models and large-scale learning frameworks for manipulation and autonomous planning.
• Build full-stack training pipelines for robotics with multi-modal inputs and reproducibility.
• Advance data collection strategies for model training and validation.
• Bridge the sim-to-real gap for embodied AI using simulation frameworks.
• Create scalable inference loops combining perception with real‑time motion control and on‑robot deployment.
• Collaborate with cross-functional teams to deploy robust and safe robotics systems.
• Present results to internal partners and at conferences.

Skills

C++

Python

English

Transformers

VLAs/VLMs

Diffusion Models

ROs/ROS2

Education

Ph.D. or MSc in CS/EE/Robotics

Tools

ROS/ROS2

Job description

Ellenco Estágios e Treinamentos in Berkeley, CA seeks a Research Scientist in Embodied AI to architect robotic foundation models and lead the design of large‑scale learning frameworks for industrial manipulation and autonomous planning. You will build end‑to‑end training pipelines for multi‑modal robotics data, bridge the sim‑to‑real gap, and develop scalable inference loops for real‑time control.

Ideal candidates have Ph.D./M.Sc.