Nomagic
AI
ResearchScientist
Neural analysis suggests this role is
optimal for Mid+ candidates.
“Research Scientist at Nomagic. Skills: Machine learning, Robotics, Large-scale multimodal model training, Foundation Models, Pretraining, Post-training, Deep learning architectures. Build, train, and deploy foundational models. Define model architectures, objectives, and training curricula across multimodal robotic data”
What You'll Achieve.
Bring our fleet from a classical control stack to generalized AI mastery; Make general models useful, controllable, and safe in the real world; Measure actual robot performance and failure modes far beyond the limits of simulation; Seamlessly bridging the gap between foundation model outputs and physical-world outcomes
Industry & Context.
Solving real problems; Bridging the gap between world-class ML research and industrial-scale robotic execution; Turning raw deployment logs into generalizable capabilities; Improve Physical Robustness; Steering robot behavior
Relocation package, Flexible working hours, English-speaking environment
What They're Looking For.
Must Have
Deep research and practical experience at the intersection of machine learning, systems engineering, and physical robotics, Experience designing, training, and fine-tuning large-scale deep learning architectures (VLMs, VLAs, RL, RLHF, Imitation Learning), ideally with policies deployed and validated on real hardware, Deep learning framework fundamentals (PyTorch/JAX), Comfort debugging at every layer of the stack, Care about empirical rigor as much as raw iteration speed, Comfort working hands-on with hardware, Understand the robotics full stack (perception, controls, state estimation), Care deeply about evaluation and failure analysis when software meets the physical world, Ability to move seamlessly between theoretical design and physical implementation, Prefer execution, rapid iteration loops, and real-world robustness over academic purity
What You'll Do.
and deploy foundational models
Define model architectures
and training curricula across multimodal robotic data
Develop scalable data mixtures and sampling strategies
Run rigorous ablations to understand scaling laws
optimization dynamics
and large-model failure modes
Collaborate closely with ML Infra to push cluster utilization and throughput
Explore fine-tuning recipes to make general models useful
and safe in the real world
Develop cutting-edge methods for improving real-world reliability
handling out-of-distribution edge cases
and steering robot behavior
Design evaluation frameworks and lightweight physical setups
Analyze real-world evaluation results to guide the overarching research direction
How You'll Work.
Team & Collaboration
Collaborate closely with ML Infra
Full Job Description
## Description Do you believe the path to general-purpose physical AI runs through noisy, real-world factory deployments? Are you excited by the challenge of turning the classical robotic stacks into the foundational training data for physical AI? Do you want to bridge the gap between world-class ML research and industrial-scale robotic execution? If your answers are yes, we should talk. At Nomagic, we are executing a humble pivot for general-purpose physical AI. We believe that physical AI is fundamentally a knowledge transfer problem - we are leveraging the "internet data" of robotics - massive deployment logs from real systems operating in production environments - to bootstrap our efforts. We are looking for Research Engineers who will help us to build, train, and deploy foundational models that bring our fleet from a classical control stack to generalized AI mastery. ## Offer essentials Play with real robots, solving real problems, every day. Relocation package. Flexible working hours. English-speaking environment. ## Here is why we love this job ourselves, and hope you will enjoy it too We combine world-class research with top-notch engineering and apply it to solve real problems. Much of this data already exists. We have robots in production at scale. We aren't waiting for datasets to be collected; the byproduct of our machines doing useful work is being created right now. We measure what matters. We test our code in unit tests, simulations, and directly on real robots. Grounding our models in deployment allows us to truly measure performance, not just offline metrics. High leverage, high impact. We’re still a highly focused team. If your architectures and training curricula improve our agents, you directly change the economics of the company. World-class peers. Our team has built Google Warsaw, unicorn startups, led research in DeepMind, tested rocket engines, and worked at top companies like Nvidia and ByteDance. Now, we are shaping the reality of Physical
Applying for this Research Scientist role?
Most applicants get filtered before a human reads their resume. See if yours makes the cut.
How to Apply on Lever
- Lever uses a streamlined one-page form — apply in under 5 minutes.
- LinkedIn import works well; review parsed data before submitting.
- The cover letter field is optional but visible to reviewers — use it to differentiate.
- Referral codes from employees can significantly boost visibility of your application.
ANONYMOUS · UNFILTERED
What do employees actually say about Nomagic?
Real rants from real employees. Read before you apply.