Staff Machine Learning Engineer, VLA Training

Sourced from emploive.com

Nidus, a startup building autonomous manufacturing systems powered by AI and robotics, is hiring a Staff Machine Learning Engineer (via Rethink Recruit) to own major parts of the model development lifecycle for vision-language-action (VLA) models. You will design training recipes and model architectures, build scalable distributed training pipelines on multi-node GPU clusters, and validate learned policies on physical robots. The role sits at the intersection of ML research, large-scale systems, and robotics, working closely with researchers, robotics engineers, and data teams.

• Train and improve vision-language-action models for robotic manipulation and embodied tasks
• Develop model architectures and training recipes spanning imitation learning, behavior cloning, transformer- and diffusion-based policies, and multimodal pretraining
• Build scalable pipelines for pretraining, fine-tuning, evaluation, checkpointing, and model release
• Train models across multi-node GPU clusters while improving utilization, throughput, stability, and cost efficiency
• Design data mixtures, sampling strategies, augmentations, and curricula for large heterogeneous robot datasets
• Develop data loaders and preprocessing for synchronized video, language, robot state, actions, and other sensor modalities
• Create reproducible experimentation systems with configuration management, dataset/model versioning, experiment tracking, and regression testing