Introducing Seldon Core
Our new open-source framework for deploying machine learning models on Kubernetes.
Artificial intelligence. Real impact.
Accelerate data science
Organisations replicate effort building and maintaining machine learning infrastructure. Seldon enables you to get started quicker and easier, with low capital overheads.
Focus on improving business KPIs
Data scientists are freed up to focus on the core model creation while devops teams are able to manage deployments effectively using tools they understand. Seldon is built on industry-standard open-source technologies that work seamlessly with the tools your team are already using.
Reduce governance bottlenecks
Enterprise machine learning deployment requires complex governance processes that increase risk. Seldon enables you to release models faster with bank-grade audit trails, approvals and model explanations.
Rock-solid enterprise machine learning stack
Seldon Core is a high-performance ML serving engine based on an advanced model graph engine for deploying experiments and ensembles.
Seldon Deploy (coming 2018) integrates seamlessly with a game-changing UX for better, faster and more compliant production workflows between data scientists, data engineers and business managers.
Deploy models from any machine learning library or framework using microservices and containers
Run Seldon on your favourite cloud provider, on-premise, or fully managed by us – it’s your choice.
We built Seldon to scale with your organisation, and are comfortably serving billions of predictions each month.
Move faster with Seldon on your team
Treat Seldon as an extension of your data science team or work with us to kick start your machine learning journey. We’re here for you every step of the way, from analysis and requirements specification to scaling a machine learning pipeline in production. Our team has experience delivering up to 30x ROI for some of the world’s leading financial services and technology companies.
Watch our one-minute explainer video
Seldon Deploy, coming 2018
We are currently building a new enterprise product that helps data science teams move from R&D into production faster with more transparency and better compliance.
Seldon Deploy streamlines the data science workflow, with audit trails and approvals, advanced experiments, model explanations, and more — all served through a stunning UI. To find out more about Seldon Deploy and why we’re building it, please check out this blog post and apply for the closed beta.
Seldon Core was open-sourced in January 2018 and has been designed from the ground up to work seamlessly with Seldon Deploy.