Senior Applied Researcher

About Bellboy Robotics

We are building the future of work with robotics. Bellboy Robotics is a VC-backed company developing intelligent mobile robots designed to automate complex, unstructured industrial and commercial environments. We are moving beyond rigid, lane-based automation to create robots that can truly perceive, reason, and act in the real world. Our mission is to make robots and people work as one team in real-world environments.

We are already deployed and commercially operating across multiple sites in world-class hotels throughout Europe, with live operations today and an active expansion roadmap across the region. Our seasoned team sits at the intersection of hardware, AI, and software - bringing together engineers from companies like Apple, Amazon, Microsoft, Mobileye, and SpaceIL.

The Role

We are looking for a Senior Applied Researcher to join our Applied AI team in Tel Aviv. You will drive the development of advanced perception and behavior models that allow Bellboy robots to interpret and act within complex, unstructured environments.

This role sits at the intersection of cutting-edge research and real-world deployment. You will build models that perform not just in simulation or on benchmarks, but on robots operating every day in live customer environments. Because our work involves hands-on interaction with physical hardware, this is a hybrid role (2-3 days a week in the office), allowing you to collaborate side-by-side with our robotics and systems engineers while maintaining flexibility.

Key Responsibilities

• Train and fine-tune Vision-Language-Action (VLA) models to drive complex robot behavioral modeling.
• Apply and adapt modern VLMs such as Qwen-VL, Grounding DINO, and SAM to embodied robotics tasks.
• Develop and refine 3D perception models for understanding structure, geometry, and spatial relationships in real environments.
• Apply model distillation, quantization, and other optimization techniques to ensure efficient inference directly on robot compute.
• Combine established computer vision models (e.g., ResNet, YOLO) with modern vision-language approaches to tackle open-world perception and grounding.
• Own the research-to-production lifecycle, from rapid exploration and prototyping to deployment on robots in the field.
• Collaborate closely with robotics, systems, and software teams to ensure model reliability, safety, and strong real-world performance.

Requirements

• MSc or PhD in Computer Science, Computer Vision, Robotics, or Deep Learning (or equivalent industry experience).
• Proven track record of building, adapting, and fine-tuning classical and deep learning Computer Vision models.
• Strong software engineering skills in Python and PyTorch, with experience writing clean, modular, production-ready code.
• Deep understanding of model architectures, representation learning, generalization, and real-world robustness.
• Experience bridging the gap between research and production; you are driven to see your code work in the physical world.
• Curiosity, urgency, and a drive to build AI systems that operate reliably in dynamic environments.

Bonus Points

• Hands-on experience with modern Embodied AI paradigms, including VLM/VLA architectures (e.g., GR00T) and Motion Policy Learning (e.g., Diffusion Policy).
• Experience in robotic perception for manipulation, navigation, or real-time edge deployment.

Why Join Us?

• Work on the cutting edge of Physical AI and Embodied Intelligence.
• See your algorithms deployed on robots solving real problems in the physical world.
• Collaborate with a high-performance team of researchers and engineers in the heart of Tel Aviv.
• Be a core part of a company defining the next generation of industrial robotics.