Just drawing circles with #gitbob 😊

#gitbob can also assist manuscript writing while logging which part of the text it modified and who reviewed it afterwards. For example: #gitbob wrote the abstract of the preprint and here is how I instructed it:

With #gitbob, we can even do basic data analysis and plotting tasks using English language directly in Github issues and see the results immediately.

New in the #stackview GitHub repository: .#gitbob answers support requests. 🥳.

I just taught #gitbob how to plot 😃.

You can give #gitbob custom instructions to achieve this. E.g. here are the instructions for the reviewer:

The word cloud #gitbob made from the #I2K2024(virtual) program 😃

🚨 #preprint alert 🚨 ."Towards Transparency and Knowledge Exchange in AI-assisted Data Analysis Code Generation": Using direct LLM-integration in discussions about Github issues and pull-requests, a.k.a. #gitbob. Let me guide you through it in a short 🧵.

If you're interested in #BioImageAnalysis code generation and how to use #biabob and #gitbob, join my session at #I2K2024 virtual coming Monday 15:00 CEST and also check out the other amazing sessions! 🔬🖥️🌐🔓.

I just updated the #gitbob #preprint: It now comes with a graphical abstract outlining how multiple humans can interact with multiple LLMs in a persistent discussion thread about #DataAnalysis tasks. 😎

The new #gitbob 0.10.1 also runs in Gitlab! Thanks to Volker Hilsenstein for proposing and thanks to gpt-4o which roughly knew how to implement this 🫠.Feedback about the installation instructions are welcome!.

Just playing tic-tac-toe against #gitbob 🕹️😄.Watch the game live 👀.

It turns out #gitbob wrote 15% of its own code 🤩 Excited to see how this will develop in the future 😃.

I asked #gitbob to redraw what it sees in an image and I must say: I couldn't have done it better myself 😅.

As #gitbob is just a GUI for modern LLMs, you can ask it basically any question, for example about #ResearchDataManagement 🎉 and it will respond. ➕You can conserve how things were done & decided. ➕Others can learn with you how to prompt LLMs. Example:

I managed to make two #gitbob's with different _personalities_ (an Editor and a Reviewer) discuss my changes in a github pull-request.😅.Ofc, an academic example; and maybe a way towards multi-agent #AI systems with humans in the loop😃.

Well designed #AI-based assistants may become better with every new LLM that comes out, for example today: we can just ask #gitbob to use claude-3-5-sonnet-20241022 and it just works 🥳.

@gitb0b Gm GitBob! 🐳.

@gitb0b Gm GitBob 👊☀️.