Research Scientist - Physical AI for Real-World Robotics

Almaz Capital; Warszawa; On-site; PLN 350,000 - 550,000; Full time

Benefits offered by this job

Relocation package

Flexible working hours

English-speaking environment

Play with real robots

Job summary

Nomagic is seeking a Research Scientist to advance foundation models for physical AI in industrial robotics. You will blend ML research with robotics systems and scale training across multimodal data collected from real deployments.

You will design architectures, run ablations, and collaborate with ML Infra to translate research into robust real-world robot performance in production environments. English-speaking environment and relocation package offered.

Qualifications

• Deep research experience at the intersection of machine learning, robotics, and systems engineering.
• Proven track record in designing, training, and validating large-scale deep learning architectures.
• Fluent with PyTorch/JAX and practical debugging across the stack.
• Hands-on robotics experience and a pragmatic mindset for real-world deployment.

Responsibilities

• Explore foundation models and pretraining for multimodal robotic data.
• Design architectures and training curricula; perform ablations and scaling studies.
• Collaborate with ML Infra to optimize distributed training.
• Guide post-training adaptation and real-world evaluation.

Skills

ML research

Robotics systems

Engineering fundamentals

Hands-on robotics

Experimentation

Communication

Tools

PyTorch

JAX

Job description

Nomagic is seeking a Research Scientist to advance foundation models for physical AI in industrial robotics. You will blend ML research with robotics systems and scale training across multimodal data collected from real deployments.

You will design architectures, run ablations, and collaborate with ML Infra to translate research into robust real-world robot performance in production environments. English-speaking environment and relocation package offered.