Be as effective as AI
engineers at Google
and Facebook

Workflow tools to help you experiment, deploy, and scale.
By data scientists, for data scientists.

hero dashboard with experiment and deployment graphs
1

Experiment

2

Deploy

3

Scale






Experiment

Power up your existing workflow with model versioning, environment setup, and experiment reproducibility.
Available now in Datmo Open Source.

collaborate on model versions with peers

Setup up your environments in seconds

Use one of our preconfigured environments, or bring your own. Setup using a single command:
$ datmo environment setup

  • Use any number of files to define your environment:
    requirements.txt, Dockerfile, .bashrc, and more
  • Instantly spin up your favorite workspace with
    $ datmo notebook or $ datmo rstudio

collaborate on model versions with peers

View your experiments in one place

Compare metrics, model configurations, and task runs directly in your terminal with:
$ datmo run ls

  • Easily reproduce any of your previous experiments with the
    $ datmo rerun command

  • Local dashboard for easy searching, filtering,
    and sorting of experiment logs

Deploy

Turn your local models into dynamic cloud services and endpoints. Any cloud, any machine learning framework.
Available now in the Datmo product.

collaborate on model versions with peers

Easy deployments for new algorithms and experiments

Don't waste your data scientists' time writing ops code. Go from local script to deployed service with:
$ datmo deploy


  • Flexible and high performance serving system for
    machine learning models in production as APIs

  • Perform A/B testing over multiple algorithms
  • Dashboard to monitor and manage all services

Scale

Reliable scaling of infrastructure services for experiments, training, and inference.
Available now in the Datmo product.

collaborate on model versions with peers

Scale out inference code on any framework or cloud platform

Cloud infrastructure prices fluctuate constantly. Datmo enables you to seamlessly deploy cross-cloud for the lowest possible rate.


update your model versions

Autoscale based on latency and
throughput needs

Datmo ensures you're only ever paying for the resources you need. Set automated triggers and thresholds specific to your business use case.

Partners

scikit-learn, keras, tensorflow, and any other framework. Amazon web services, google cloud form, microsoft azure, and any server or cloud

Join our newsletter for updates: