top of page

MLOps for Teams On a Mission
Scales from Laptop to Cloud, On-Prem & Cross- or Hybrid-Cloud
Enabled for Azure, AWS, Kubernetes and Edge Deployments

By its unique design based on widely used and state-of-the art  components in the distributed computing stack, omega-ml is easily deployed to any cloud. It works best when your cloud uses Kubernetes, however bare-metal, Docker or other container technology supporting Python, RabbitMQ and MongoDB will work.

​

With the Open Source core, omegal|ml is readily deployable by any DevOps team familiar with Docker. The Commercial Edition is readily enabled for Kubernetes.

​

For on-premise deployments or if you need enterprise-ready security and do not want to manage the complexity of a large-scale data & compute cluster, the omega-ml Commercial Edition is available. 

​

Setup is a breeze: omega-ml Commercial Edition is readily configured to support any Kubernetes cloud. Deployed using our GitOps-enabled Cloud Manager, omega-ml is ready to run AI & ML workloads at any scale.

​

The built-in JupyterHub, AppHub and the serverless lambda-style Runtime Worker services are instantly ready to use, providing Notebooks for straight-forward team collaboration and interactive computing at scale. All while integrating your DBMS and existing data science pipelines to run from the same easy-to-use API.

​

Fun Fact: Unlike most vendors we do not license by number or size of compute nodes - unlimited compute capacity means your license cost is known in advance and stays fixed.

DELIVER BUSINESS VALUE
HELLO PLATFORM ENGINEERS

Leverage the Cloud, no Vendor Lock-in

All globally leading cloud providers offer their version of Ready-To-Go AI Services. However, they come at the cost of unavoidable vendor-lock in.

Don't let this happen to you, use the vendor-independent omega-ml platform to stay flexible.

​

Make Platform Engineering & DevOps Happy

The omega-ml MLOps platform provides a cost-effective and scalable approach to run AI and ML workloads - all while keeping infrastructure, data and processes under control. No Lock-In. Your Data Center. Ready today. 

​

Enable AI Teams To Deliver Business Value

Analytics teams need MLOps to leverage the full potential of AI and machine learning. There is no value in building this capability in-house. Use omega-ml and start delivering.

An Cloud

AI Delivery Platform

Open & Flexible Integration from Development Lab to Production

omega-ml comes with batteries included, however new requirements are not a problem: Alternate data sources & sinks, third-party applications and services can easily be added thanks to omega-ml's flexible multi-layer and microservices archicture.

The following extension points are available:

feature store and model repository

STORAGE BACKEND

Extend what  objects can be stored and retrieved

pre-deployed model runtime

RUNTIME BACKEND

Add any compute backend to run any models through the same API

any cloud or on-premises

COMPUTE & DATAFRAME

Add processing options such as AutoML or Model Visualization

Architecture

Works with any machine learning framework

# Python and R frameworks

​

# — scikit-learn learn models and pipelines

pipl = Pipeline([…])om.models.put(pipl, ‘forecast’)

om.runtime.model(‘forecast’).predict(‘datax’)

​

# — R models

om$models$put('r$model', 'mtcars-model')
om.runtime.model(‘mtcars-model’).fit(‘datax’)

​

# — MLflow projects

lflow_path = 'mlflow:///path/to/mlflow/project' meta = om.scripts.put(mlflow_path, 'myproject', kind='mlflow.project')

om.runtime.script(‘myproject’).run()

Examples
bottom of page