AI-Assisted programming for research software

[yabellini]

Who is the audience?

Who might want to know this? What other groups or organizations might be interested?

• Scientific software developers already using AI or are interested in learning more about it.
• There are concerns about the quality of code AI produces, including hallucinations from LLM-based tools.
• There are concerns about the ethical aspect of developing/using these tools.
• Teachers of courses that have programming.
• Who else?

Why is this important?

The tools and resources are there, and people are using them or wondering which and how to use them. Research shows that Developer's "AI skill threat" is genuine, and creating a learning environment reduces the feeling of threat. (More info here: https://www.pluralsight.com/resources/blog/software-development/ai-in-software-development-study-results)

What should be covered?

• Use cases
• Different models/tools

Suggested speakers or contributors

• Curious to hear who you would like to talk to

Resources you would recommend to the audience

From rOpenSci:

• We have the co-working session AI in/for R hosted by @joelnitta

Others:

• Jon Udell series of blog posts on using AI to code
• Simon Willison's Webblog
• Posit AI Blog
• Learn AI-assisted Python Programming: With GitHub Copilot and ChatGPT by Leo Porter and Daniel Zingaro.
• What else?

[SunnyTseng]

I am interested in this topic and would love to hear a discussion about it. 😍 🤟 🚀

[kaijagahm]

I am also interested in a discussion on this topic!

[erikaris]

interested in this topic.

[nmorandeira]

I'm interested!

[schneiderpy]

Me too ...

[mpadge]

Self-promo here, but there's an interview on the ethics of code assistants here (... with me :wink: )