PyAmsterdam meetup @ Aidence offices ===================================== Hello everyone! We are coming back in June thanks to Aidence who will provide food and drinks for us 💪! About Aidence ------------- Aidence [1] is a company that uses Python to develop an AI-powered clinical applications for the lung cancer detection. Using deep- learning, Aidence helps to automate lung nodule detection and quantification on chest CT scans. Schedule -------- +-------+----------------------------------------------------------------------------------------------------------+ | 18:00 | Doors open! let's share some food and drinks thanks to our host. | +-------+----------------------------------------------------------------------------------------------------------+ | 19:00 | First talk: "Creating Python Code with Copilot" -- Peter Collewijn | +-------+----------------------------------------------------------------------------------------------------------+ | 19:30 | Second talk: "Writing bug-free Python code: a journey from tests to formal verification" -- Gram Voronov | +-------+----------------------------------------------------------------------------------------------------------+ | 20:15 | Lightning talks? | +-------+----------------------------------------------------------------------------------------------------------+ | 20:30 | Networking | +-------+----------------------------------------------------------------------------------------------------------+ | 21:00 | Closing time | +-------+----------------------------------------------------------------------------------------------------------+ Creating Python Code with Copilot --------------------------------- About Peter Collewijn Peter is a business consultant, software tester with a focus on testing of AI based applications. He also gives presentations on this topic. He is member of the working group Testing and AI of TestNet and promotes the AI for Business and Government program of the NL AI Coalition. He uses Python for testing and creating Machine Learning models. Abstract ~~~~~~~~ Github calls Copilot an AI pair programmer that helps you write code faster and with less work. See their site: https://copilot.github.com/. Copilot is available as an extension for Visual Studio Code and is based on the OpenAI Machine Learning model GPT-3. This is one of the most advanced Machine Learning models available today. In this presentation I spend a few minutes on the basics of AI and Machine Learning. After that I will show some possibilities of the GPT-3 model in a Jupyter Notebook. This will help you understand how Copilot operates. And then I switch to Visual Studio Code for a demo of Copilot. You will see how Copilot interacts with the user, the level of understanding of the user requests and the quality of the code generated. If we have time, it will be possible to have Copilot create code based on suggestions from the audience. Writing bug-free Python code: a journey from tests to formal verification ------------------------------------------------------------------------- About Gram Voronov Author of 100+ open-source projects, most of them Python. https://orsinium.dev Abstract ~~~~~~~~ The talk is my 5 year journey into the world of writing safe and reliable code. The result of this journey is Deal, a library that is easy to use, but under the hood it hides an enormous machine of test generation, static analysis, and formal verification. Do you want to know how it works? Links ----- [1] https://www.aidence.com/