Staff ML Engineer — Vision-Language for Robotics

Nidus; Northern, New York (KY, NY); Hybrid; USD 120,000 - 180,000; Full time

Job summary

Nidus Technologies is seeking a Staff Machine Learning Engineer to lead development of vision-language-action models powering our autonomous robots. You will own dataset construction, training recipes, and scalable training infrastructure, collaborating with researchers, robotics engineers, and data teams to deploy robust robot capabilities.

This hands-on role sits at the intersection of ML research, large-scale systems, and robotics, requiring strong communication and a track record of

Qualifications

• Five+ years of professional experience in ML, robotics, computer vision, or related field.
• Strong experience training modern deep learning models using PyTorch/JAX or similar frameworks.
• Experience with transformer-based and multimodal models, or learned control policies.
• Experience building and operating distributed training pipelines on multi-GPU/ multi-node clusters.
• Experience with large-scale datasets, high-throughput data loading, experiment tracking, and reproducible ML workflows.
• Strong understanding of optimization, model architecture, data quality, evaluation methodology, and experimental design.
• Ability to debug failures across the full stack from data to robot behavior.
• Bachelor's degree in CS/ML/Robotics or equivalent practical experience.

Responsibilities

• Train and improve vision-language-action models for robotic manipulation and embodied tasks.
• Develop model architectures and training recipes for multimodal policies.
• Build scalable pipelines for pretraining, fine-tuning, evaluation, and model release.
• Train models across multi-node GPU clusters, improving utilization and cost efficiency.
• Design data mixtures, augmentations, and curriculum strategies for large robot datasets.
• Develop data loaders and preprocessing for synchronized video, language, and robot state data.
• Create reproducible experimentation systems with versioning and tracking.
• Define offline and on-robot metrics for success, generalization, and safety.
• Build tools to inspect trajectories and diagnose training issues.
• Run experiments on physical robots to guide model and infrastructure improvements.
• Collaborate with data-collection teams to improve coverage and data quality.
• Translate research into production-quality systems for repeated deployment across robots.
• Make technical decisions across architecture, data, and compute, communicating tradeoffs.
• Mentor engineers and establish engineering practices for ML codebase.

Skills

Deep learning

PyTorch/JAX

Transformer models

Distributed training

Multimodal models

Experiment tracking

Communication skills

Debugging

Education

Bachelor's degree in CS/ML/Robotics

Tools

ROS/ROS 2

MuJoCo

Gazebo

PyTorch

Job description

Nidus Technologies is seeking a Staff Machine Learning Engineer to lead development of vision-language-action models powering our autonomous robots. You will own dataset construction, training recipes, and scalable training infrastructure, collaborating with researchers, robotics engineers, and data teams to deploy robust robot capabilities.

This hands-on role sits at the intersection of ML research, large-scale systems, and robotics, requiring strong communication and a track record of