09-15, 13:40–14:10 (Europe/Amsterdam), Foo (main)
ChatGPT is a fantastic assistant, but it cannot do everything yet. For example, it cannot automatically manage my calendar, update my to-do list, or do anything that requires it to perform actions. However, what would it take to make this a reality? I decided to put it to the test by allowing ChatGPT to manage my to-do list for me.
During this presentation, I will tell how I gave ChatGPT access to my to-do list. Along the way, I will introduce you to the concepts behind LLM-based agents and how they work. Of course, I will also give a demo of the final result. After this demo, we will dive into clever engineering solutions and tricks I discovered to solve problems such as handling hallucinations, parsing actions, etc.
This talk is for people who want to learn how to build their first LLM-based agent. Familiarity with Python, PyDantic, and LMMs is nice during this presentation but not essential. As long as you love overengineered solutions to a basic to-do list, you will like this presentation.
During the presentation, we will discuss things such as:
- How to give ChatGPT access to your ToDo(ist) list?
- What are LLM agents?
- What is the REACT framework?
- A demo of the agent I built to manage my to-do list.
- Implementation tips and tricks to make the agent work better.
The repo can be found here:
github.com/j0rd1smit/todoist_react_agent
No previous knowledge expected
Hi! My name is Jordi Smit. I’m deeply passionate about software engineering, data science, and automation. Nothing makes me happier than creating software that helps humans by automating a tedious and manual-intensive part of their job. Therefore, I love discussing data science since this field has opened the door to many new kinds of automation. However, data science solutions often stay stuck at the proof of concept level. To combat this issue, you also need software engineering knowledge. That is why I love the intersection between software engineering, data science, and automation.
I work as a Machine Learning Engineer at Xebia Data in Amsterdam. Here, I help companies to transform their ML-based models into production-ready applications. I love this job because it allows me to explore the intersection between software engineering and data science daily.