Skip to content

Introduction to NexCore ​

NexCore is the embodied AI skill learning platform from Lumos Robotics. It brings every stage of teaching a robot a new skill into a single platform: collecting demonstrations, processing and curating data, training models, delivering models, and deploying them to real robots.

The platform produces three kinds of assets: datasets, models, and meta-skills.

Capabilities ​

CapabilityDescriptionLearn more
Egocentric data collectionCapture human demonstrations with Lumos Ego, UMI, and other robot-free devicesEgocentric Data Collection
Teleoperation data collectionCapture on-robot skill data by teleoperating Lumos robots such as Station Lite and MOSTeleoperation Data Collection
Data upload and managementOrganize data assets by collection task, with multi-dimensional filtering and managementData Management
Data pipelines and processing jobsTurn raw recordings into structured data with pipelines built from standard operatorsData Pipelines
Annotated data generation, management, and deliveryGenerate, export, and download quality labels, time ranges, trajectories, depth, keypoints, and other annotationsAnnotated Data and Delivery
Model training and job managementTrain models on datasets using built-in model templates, with checkpoint export and continued trainingModel Training
Model artifact management and downloadTrace model lineage and download model artifactsModels and Downloads
On-robot model deploymentCloud or edge deployment with safety confirmation, on-site validation, and data feedbackOn-Robot Deployment

The Skill Learning Loop ​

NexCore workflow

  1. Collect: Bind a collection task to a specific device, record demonstrations, and upload them.
  2. Manage: Uploaded data becomes data assets, traceable to the collection task that produced them. In a curation project, you flag quality issues, define time ranges, and review each episode.
  3. Process: Egocentric data passes through a data pipeline to produce structured data. No processing step ever modifies the original data.
  4. Train: Export curation results as a frozen dataset version and use it to create a training job, which produces a model version.
  5. Deploy: Deploy the model to the cloud or an edge node, then validate and run it on a real robot.
  6. Iterate: Execution data flows back automatically. Failure cases (bad cases) are reviewed and returned to data management for the next iteration.

Every stage records versions and lineage, so for any model you can answer: which data was used, how it was processed, and which version it is pinned to.

Who NexCore Is For ​

AudienceKey questionsStart here
Enterprise and production customersCan we collect data, train, and deploy skills in our own environment?Tutorials
Research institutions and universitiesAre the data formats open? Can we export datasets and models for our own experiments?Data Formats · Downloading Models
Partners and system integratorsHow do we connect devices and deploy models on site?Device Management · On-Robot Deployment
Operations leadersHow does the platform drive productivity, and how should the team divide the work?Core Concepts · Roles and Permissions
Skill developersHow do we build new models, pipelines, and skills?Quick Start

Platform Architecture ​

LayerComponents
Devices and computeLumos-built collection and execution devices, including the Station Lite dual-arm workstation, the heavy-duty MOS mobile manipulator, and robot-free Ego and UMI devices; cloud training compute; on-site edge inference nodes
Platform toolsCollection tasks, data assets, curation projects, data pipelines, datasets, training jobs, model management, deployment, evaluation, and feedback
Model and skill assetsDatasets, models, skills
AccessWeb console and edge client

NexCore is served from a single domain, with compute and storage nodes deployed by region. Data is written to the storage nodes in its region, training jobs are scheduled onto available compute, and edge inference nodes only need outbound connectivity to the platform. For data that must remain entirely on premises, a private deployment is also available.

Design Principles ​

PrincipleWhat it means
Read-only source dataCuration and pipelines only produce metadata, instructions, and new outputs. Original data is always preserved.
Frozen versionsDatasets are frozen on export, and training binds only to frozen versions, so results are reproducible.
End-to-end lineageData IDs, dataset versions, model versions, and deployment records are linked at every step.
Open formatsDatasets export to open formats such as HDF5 and LeRobot, and model artifacts are downloadable.
Project isolationData is isolated by project and is not visible across projects.
Safety firstA safety confirmation is required before any on-robot run.

Scope ​

  • In addition to Station Lite and MOS, third-party robots can be integrated through adapters. The scope of each integration is assessed per project.
  • NexCore provides the tools and assets. Scenario-specific solutions are built jointly by customers, partners, and field application engineers.

NexCore is evolving rapidly; the latest version prevails.NexCore Docs · Embodied AI Skill Learning Platform by Lumos Robotics