DataKRK #39 - AI Infrastructure Trends, Snowflake Zero Copy Cloning & YOLOv7

Let’s meet at the next Data KRK session! Check out the details below. ⤵️

When: Tue, 21 March 2023, 6:00 PM
Where: Hevre - Meiselsa 18, 31-058 Kraków

On the agenda:

🎙️ Lightning talk - I read 100 pages of AI Infrastructure 2022 Report so you don't have to by Zbyszek Królikowski

In mid-2022, the AI Infrastructure Association, a group of ML market leaders, published a report on the current state and possible future of ML engineering tools. The text included insights on companies, changes in the solutions used and the general future of data science, data and ML engineering. However, the report is quite long, so I decided to read it for you and share my insights and observations.

🎙️ Short-lived test environments with Snowflake Zero Copy Cloning by Tomasz Kostyrka

Automating the deployment of data warehouse projects can be a complex task. One of the biggest challenges when working with this kind of solution is the inability to quickly and completely test changes before deploying them to production environments. In most cases, conducting such tests may require setting up a dedicated environment, as close as possible to production, and running ETL/ELT processes on it. As it may not always be feasible to set up a copy of a warehouse with terabytes of data, we might be forced to use workarounds and simplifications, such as shared testing environments or limited volume/sample datasets. In this session, we will explore how the Snowflake platform's Zero Copy Cloning and Time Travel functionalities, in combination with a well-designed CI/CD process, can solve these challenges and help us properly verify our changes before we hit the "GO LIVE!" button.

🎙️ 3D Pose Estimation with YOLOv7 by Piotr Skalski

Are you ready to take your sports analysis to the next level? Look no further! In this talk, we will dive deep into the exciting world of 3D pose estimation using multiple cameras and the powerful YOLOv7 model. From detection to post-processing, calibration to visualisation, I'll walk you through every step of the process and provide you with the professional insights you need to improve your analysis. But don't worry, this talk won't be all work and no play - I promise to add some humour to keep things interesting.

🎙️ Networking over a beer

Speakers:

Zbyszek Królikowski | ​​Machine Learning Engineer @ VirtusLab

Zbigniew is a Software Development Engineer at VirtusLab. He works on bringing ML recommendations and re-ranking models onto production, both in private and public clouds.

Tomasz Kostyrka | Senior Data Platform Consultant @ Elitmind

He has been working in the Data Management area for 10 years and despite everything he has seen so far, he still likes it. A big fan of automation, willingly involved in topics related to DevOps/DataOps. Privately, a tired father longing for a good sleep.

Piotr Skalski [SkalskiP] | ML Growth Engineer @ Roboflow

Hi there! My name is Piotr Skalski, and I am a Computer Vision Engineer with over 5 years of experience in the field. My primary focus in my work is on detection, segmentation, and tracking, but my passion for sports is what truly sets me apart. I am always looking for ways to connect sports and computer vision, and I am constantly exploring new and innovative ways to use technology to enhance sports analysis.

Remember to mark your attendance and see you on 21st March! 👋