Automate the Data Science Pipeline with Serverless Functions

serverless functions
Nuclio Serverless Functions

Nuclio is an open source and managed serverless platform which allows developers to build and run auto-scaling applications, managing resources efficiently and cutting time to deployment


The fastest platform running up to 400,000 function invocations per second


Enables simple debugging, regression and a multi-versioned CI/CD pipeline


Supports a variety of open or cloud-specific event and data sources with common APIs


Is portable across low-power devices, laptops, on-prem and multi-cloud deployments


Enables automated deployment to production in a few clicks from Jupyter notebook

Learn more about serverless functions here
Automate Data Science From Development to Production
See examples here
Take Off: Get Started in 60 sec!

The simplest way to explore Nuclio is to run the nuclio dashboard (you only need docker):

docker run -p 8070:8070 -v /var/run/docker.sock:/var/run/docker.sock -v /tmp:/tmp nuclio/dashboard:stable-amd64


Browse to http://localhost:8070, deploy one of the example serverless functions or write your own. The dashboard, when running outside an orchestration platform (e.g. Kubernetes or Swarm), will simply be deployed to the local docker daemon.

The Getting Started With Nuclio On Kubernetes guide has a complete step-by-step guide to using Nuclio serverless functions over Kubernetes.

Nice moves hero. What else you got?
Use Cases
See examples here
Choose How to Work with Nuclio

Self Hosted

Free, Open Source
  • High Performance (~300k events per second)
  • Kubernetes Support
  • Management Dashboard
  • GPU Support
  • Multi-Tenancy
  • Basic Monitoring and Logging


Enterprise Grade
  • High Performance (~300k events per second)
  • Kubernetes Support
  • Management Dashboard
  • GPU Support
  • Multi-Tenancy
  • Advanced Monitoring and Logging
  • Auto-Scaling and Scaling to Zero
  • Security – Authentication and Authorization, LDAP Integration
  • Resource Enforcement
  • Integration with ML tools: Spark, Tensorflow, Jupyter and more
  • Multi-Model Database with Time Series, Key-Value, SQL, Columnar
  • Quick Model Deployment From Jupyter Notebook to Serverless Functions
  • Shared Volume Across Functions
  • 24×7 Supprt
Nuclio Supports



Access documentation here

Presentations and Videos

Kelsey Hightower Tweets, Oct 30, 2018
Asaf Somekh interviews Kelsey Hightower at Serverless NYC

The Future of Serverless, Oct 30, 2018
Yaron Haviv at Serverless NYC

TheCube, KubeCon + CloudNativeCon EU, May 2, 2018
Yaron Haviv on TheCube

CNCF Webinar Series – Taking Serverless and AI to the Next Level
Taking Serverless and AI to the Next Level

SolarWinds Innovative Summit Presentation
CNCF, State of Serverless & Project Nuclio

TGI Kubernetes 022: Nuclio serverless system

How to Go Serverless to Enable Faster and Simpler Analytics

Running High-Speed Serverless with Nuclio

Nuclio Overview

CNCF Serverless Working Group, Oct 1, 2017
iguazio Presents Nuclio

TheCube, BigData NYC, Sept 27, 2017
Yaron Haviv on TheCube