Full Stack Data Science
Lean AI is a full stack data science platform, and a single application for the entire machine learning lifecycle.

Built in collaboration with:
Lean AI
Lean AI full stack data science is an open source initiative initiated and maintained by Scaleout. The purpose of the stack is to provide a highly flexible, cloud native open toolkit for full stack data science projects.

We integrate leading open source components covering all stages from data ingestion and transformation, feature extraction, model definition, training and evaluation, to deployment, inference and monitoring with infrastructure automation on top of Kubernetes.

Lean AI full stack data science will continue to evolve with the open source DevOps and ML community, and through active development by Scaleout.

Github repository: https://github.com/leanaiorg/leanaistack
Key Features
  • Machine learning framework agnostic. Supports most machine learning frameworks out of the box and with open APIs support extension for additional frameworks.
  • Continuous analytics. Support for automating pipelines from data ingestion to model deployment and monitoring.
  • Workflow orchestration. Automate your workflows to handle tasks such as grid searches or active learning pipelines.
  • Simple deployment. Run anywhere you can run Kubernetes, hosted on-prem, in the cloud or on your laptop.
Stack Overview
What problems are we solving with the Lean AI solution?
  • Deals with the complexity of integrating best-of-breed open source toolchains for machine learning and DevOps into a full stack data science solution.
  • Avoids vendor lock-in and provides a lot of flexibility in deployment, on-premise, in any major cloud provider, or at the edge.
  • Enables an agile/lean approach to AI projects, built with the aim to help cross-functional and full-stack teams deliver production solutions rapidly.
  • We follow closely the evolution of the cloud native landscape and make sure that the best open source tools for each task are used.
Product categories
Our solution consists of a complete data scientist workspace including tools for ingestion, experimentation, training, validation and visualization.

It covers lifecycle management of models with versioning, annotations, storage and distribution. Our out of the box solution packages models and metadata in containers for easy distribution and scaling.

The model serving solution utilizes best of breed solutions for serving models for different needs. We can deploy models on serverless infrastructure or container orchestration platforms such as Kubernetes. Through such solutions we provide scaling, load balancing and enable customizable and versatile model serving solutions for production.
ML studio
Portal, Dashboards, Notebooks, User and project management
ML pipelines
Workflow, Signals, Scaler
Model management
Repository, Registry
Deploy pipelines
Integration, distribution, Custom CI/CD
Dataset management
Repository
Model serving
Traffic management, Authentication / Authorization, Policies, Endpoint Registry, Monitoring / Logging
Tools and libraries
Lean AI full stack data science is building on top of the best available open source tools and components.
Contact
If you want to learn more about Lean AI and full stack data science, visit the Github repo or get in touch for a demo!
Phone: +46(0)18-7770303
E-mail: contact@scaleoutsystems.com

We are located at Epicenter in Stockholm and Juvelen in Uppsala.