Tech Talks #77

This talk covers practical tips for optimizing tensor operations in PyTorch. Learn how to speed up your code using efficient indexing, broadcasting, in-place operations, jitting, and proper device management to get the most out of your CPU or GPU.

Have you ever deployed to production?
Is your project connecting to a database?
Does this database need some changes in the structure?
Have you just ignored what happens with your production between the time you execute DB migrations and restart workers with new code?

If you answered "YES" to all those questions, this presentation is just for you!

I will teach you how to deploy your DB changes with zero downtime*!

Discover how to plan your code changes and deployments to ensure seamless access to your database—without any risk of your application momentarily falling out of sync. In this session, you'll learn techniques that prevent those brief but critical moments when your code might be a step ahead or behind your database schema. We'll also explore the broader benefits of this approach, including improved stability, smoother rollouts, and enhanced user experience.

* This isn't true zero-downtime really, but bear with me! More on the presentation!