Chronon – Open Source Data Platform for AI/ML

Abstract: 

Chronon is an open source real-time data platform for AI/ML, developed and maintained by Airbnb and Stripe. It is responsible for computing, backfilling, and serving feature transformations, with a strong emphasis on real-time computation, scalability and consistency.

This talk will go over the business impact of the platform, as well as provide a technical overview of the architecture and implementation. It would be interesting to anyone who is thinking about deploying data solutions within their organization to better support AI/ML workflows.

For business impact, use cases such as anti-fraud, personalization, and customer support will be explored. These use cases span both predictive ML as well as applications of generative AI solutions. Context will be given on these use cases, and why ML practitioners in these areas often hit issues when dealing with data.

During the technical overview, the talk will cover details of the offline compute engine, largely orchestrated in Spark. It will also cover the online side of data computation, including the lambda architecture implemented by Chronon under the head, with implementations on both Spark Streaming and Flink. During this section, there will be an emphasis on scalable computation of complex time-windowed aggregations. It will also cover the online APIs that Chronon offers for data fetching, and illustrate the end-to-end ML flows that one can build using these components.

The talk will also cover some of the details of the open source roadmap, and offer insight on how to use the project within your organization. It will leave listeners with a clear sense of what value the platform can deliver, and what next steps to take if they are interested in evaluating it.

Bio: 

Varant is an ML Infrastructure engineer at Airbnb, where he focuses on accelerating the development and deployment of models through better infrastructure. He is one of the authors of the recently open sourced project Chronon. Prior to Airbnb, he worked as a Forward Deployed Engineer at Palantir, where he was similarly focused on helping organizations make better use of their data.

Open Data Science

 

 

 

Open Data Science
One Broadway
Cambridge, MA 02142
info@odsc.com

Privacy Settings
We use cookies to enhance your experience while using our website. If you are using our Services via a browser you can restrict, block or remove cookies through your web browser settings. We also use content and scripts from third parties that may use tracking technologies. You can selectively provide your consent below to allow such third party embeds. For complete information about the cookies we use, data we collect and how we process them, please check our Privacy Policy
Youtube
Consent to display content from - Youtube
Vimeo
Consent to display content from - Vimeo
Google Maps
Consent to display content from - Google