O’Reilly and TensorFlow teamed up to present the first TensorFlow World last week.

It brought together the growing TensorFlow community to learn from each other and explore new ideas, techniques, and approaches in deep and machine learning.

Presenters in the keynote:

  • Jeff Dean, Google
  • Megan Kacholia, Google
  • Frederick Reiss, IBM
  • Theodore Summe, Twitter
  • Craig Wiley, Google
  • Kemal El Moujahid, Google

Container startup time is the price of admission, but startup time is also an inhibitor for many serverless scenarios.

Steve Lasker joins Lara Rubbelke to show how Project Teleport eliminates the download and decompression of layers by directly mounting expanded layers, potentially unlocking many new scenarios.

Related Links

Practical Engineering explores a scalable (and unexpected) means of mass energy storage.

Electricity faces a fundamental problem that comes with pretty much any product that’s provided on-demand: our ability to generate large amounts of it doesn’t match up that closely with when we need it. The storage of electricity for later use, especially on a large scale, is quite challenging. That’s not to say that we don’t store energy at grid scale though, and there’s one type of storage that makes up the vast majority of our current capacity.

A quick introduction on how to think about setting up alerting and monitoring for applications in Kubernetes.

What do you get out of the box, how can you integrate more advanced metrics and add your own custom monitoring and alerting to reliably operate your application.

Resource links

O’Reilly has a a great round up post from the first ever TensorFlow World conference that took place earlier this week.

People from across the TensorFlow community came together in Santa Clara, California for TensorFlow World . Below you’ll find links to highlights from the event. TensorFlow World 2019 opening keynote Jeff Dean explains why Google open-sourced TensorFlow and discusses its progress. Accelerating ML at Twitter Theodore Summe offers a […]