< Back

Machine Learning for the Masses, TiDB & Bikesharing

April 18, 2018 6:00 PM

Machine Learning for the Masses

Albert Shau Cask

As the importance of AI has increased, it has also become easier and easier to train and use machine learned models. However, it is still largely a disconnected experience, with different tools used for data ingestion, preparation, model training, and analysis. We developed a simple, UI driven solution on top of CDAP that lets a casual user handle everything involved with ML in a single place. Data is ingested and prepared interactively. Models are trained with a few button clicks, evaluated, and automatically integrated with the rest of your pipelines, apps, and datasets, providing a unified experience to quickly go from experimentation to production.

Albert Shau is a software engineer at Cask, where he is working to simplify data application development. Prior to Cask, he worked on search systems at Box, and recommendation systems at Yahoo.

How TiDB helps Scale World's Largest Bikesharing Platform

Kevin Xu PingCap

Kevin’s talk will provide an architectural and features overview of how TiDB delivers a hybrid transactional and analytical processing database solution with its various components:

– TiDB, stateless SQL layer ( https://github.com/pingcap/tidb )

– TiKV, transactional distributed key-value storage layer ( https://github.com/pingcap/tikv )

– TiSpark, Spark plug-in for complex OLAP queries ( https://github.com/pingcap/tispark )

He will also do a deep dive into one of its in-production use cases with Mobike (one of the world’s largest dockless bike-sharing companies, serving 30 million rides per day), with details on how TiDB supports specific application scenarios.

Kevin Xu (@kevinsxu) is the General Manager of U.S. Strategy and Operations at PingCAP, the company behind the popular open source NewSQL distributed database TiDB ( https://github.com/pingcap ). He studied CS and Law at Stanford, was a design fellow at Stanford’s d.school, and served in the Obama White House.