FastAPI & QuantumBlack, AI by McKinsey

This month we are happy to have speakers from FastAPI and QuantumBlack, AI by McKinsey! Sebastian will share some tips and tricks for ML and introduce FastAPI and Juan will show how to create complex AI pipelines using Hugging Face transformers while structure them into Kedro projects.

Big thank you to Hopsworks for sponsoring snacks & drinks for the event!

–

Agenda:

17:30 - 18:00: Doors open
18:00 - 18:10: Welcome
18:10 - 18:40: FastAPI Presentation
18:40 - 19:10: Snacks & Drinks
19:10 - 19:40: Who needs ChatGPT? Rock solid AI pipelines with Hugging Face and Kedro
19:40 - 20:30: Networking

–

Presentations:

Intro to FastAPI: Tips and Tricks for ML
Sebastián Ramírez - Author of FastAPI, Typer, SQLModel, Asyncer, and more

Learn how to create an API ready for production in very little time using FastAPI... explained with memes. And learn the tips and tricks for serving Machine Learning, simple and advanced. Your API will have automatic validation, documentation based on standards, high performance, and several other features. All this, having editor support including autocompletion everywhere. You will learn how to optimize model loading, avoid common mistakes, and make development iterations more efficient.

In this talk, you will learn what FastAPI can do, and how it could benefit you, and the tips and tricks to effectively deploy ML models. You will see how to declare the data you want to receive in each request using standard Python type annotations. Including path parameters, query parameters, body payloads with JSON, etc. You will also see how to use simple, standard, Python type annotations to declare complex JSON body payloads with deeply nested structures, and get automatic data validation, serialization, and documentation. You will learn techniques to load ML models, including caching, lazy loading, and more.

Speaker Bio:
Hey! 👋 I'm Sebastián Ramírez (tiangolo), the creator of FastAPI, Typer, SQLModel, Asyncer, and other open-source tools.

I've worked with companies and teams across the world, from Latin America to the Middle East, going through Europe and USA. Always building different types of products and custom solutions involving APIs, data processing, distributed systems, and Machine Learning. 🤓

Who needs ChatGPT? Rock solid AI pipelines with Hugging Face and Kedro
Juan Luis Cano Rodríguez - Product Manager at QuantumBlack, AI by McKinsey

Artificial Intelligence is all the rage, largely thanks to generative systems like ChatGPT, Midjourney, and the like. These commercial systems are very sophisticated and powerful, but also a bit opaque if you want to learn how they work or adapt them to your needs. What happens inside the 'black box'?
Luckily there are open AI models that you can download comfortably, study without restrictions, and adjust so that they do what you want. This requires some technical knowledge, but thanks to Hugging Face's models and their ecosystem of Python libraries, delving into AI is easier than ever.
You will soon find yourself combining different models, performing different tasks, and creating complex systems. But this complexity can grow very quickly, and soon you'll find yourself with spaghetti code if you are not careful. By using the Kedro catalog and Kedro pipelines, you will be able to organize the code in no time.
In this talk you will learn how you can create complex AI pipelines using Hugging Face transformers, structure them into Kedro projects that cleanly separate code from configuration and data, and deploy them to production so they start delivering value.

Speaker Bio:
Juan Luis (he/him/él) works as Product Manager at QuantumBlack, AI by McKinsey, with a focus on Kedro, an open source data science framework. He has a decade of experience as developer advocate, software engineer, and Python trainer in several industries.

PSF Fellow since 2017, he has made significant contributions to the PyData stack, published several open-source packages, and organized the first seven PyCons in Spain. Currently he is the lead organizer of the PyData Madrid monthly meetups.

–

About the event

Date: November 23rd, 17:30 - 20:30
Location: Microsoft Reactor (Regeringsgatan 59, 111 56 Stockholm)
Directions: 2-minute walk from Hötorget.
Tickets: Sign up required. Anyone who is not on the list will not get in. The event is free of charge.
Capacity: Space is limited to 60 participants. If you are signed up but unable to attend, please let us know by November 22nd.
Questions: Please contact the meetup organizers.

–

Code of Conduct

The NumFOCUS Code of Conduct applies to this event; please familiarize yourself with it before attending. If you have any questions or concerns regarding the Code of Conduct, please contact the organizers.