# Training an LLM to generate queries from natural language prompts - GSoC 2025

**URL:** <https://aiida.discourse.group/t/training-an-llm-to-generate-queries-from-natural-language-prompts-gsoc-2025/540>\
**Category:** Developer\
**Tags:** discussion\
**Created:** [February 9, 2025, 11:55am UTC](https://aiida.discourse.group/t/training-an-llm-to-generate-queries-from-natural-language-prompts-gsoc-2025/540 "2025-02-09T11:55:07Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![jgyasu](https://yyz2.discourse-cdn.com/free1/user_avatar/aiida.discourse.group/jgyasu/32/257_2.png) [@jgyasu](https://aiida.discourse.group/u/jgyasu)\
**Post date:** [February 9, 2025, 11:55am UTC](https://aiida.discourse.group/t/training-an-llm-to-generate-queries-from-natural-language-prompts-gsoc-2025/540/1 "2025-02-09T11:55:07Z")

</div>

Hello, my name is Jigyasu. I am an electronics engineering undergraduate. I was going through the NumFOCUS accepted projects and found AiiDA there. I am interested in working on this project for GSoC '25 and I just wanted to say hi!

I have research experience in LLMs where I have fine-tuned them before, to be specific, I fine-tuned Google’s Gemma on the MedQuAD dataset. I am also collaborating with a team at UofSC to research watermarking methods for LLM-generated text.

I like open-source software because I believe it is the epitome of what human collaboration can do and in the past, I have contributed actively to `sktime`, a machine learning framework for time series and I am experienced in object-oriented programming paradigms from there.

I just set up the development environment and I will open a PR soon 🙂

---

<div class="post-metadata">

**Author:** ![jgyasu](https://yyz2.discourse-cdn.com/free1/user_avatar/aiida.discourse.group/jgyasu/32/257_2.png) [@jgyasu](https://aiida.discourse.group/u/jgyasu)\
**Post date:** [February 9, 2025, 6:05pm UTC](https://aiida.discourse.group/t/training-an-llm-to-generate-queries-from-natural-language-prompts-gsoc-2025/540/2 "2025-02-09T18:05:05Z")

</div>

I would also like to join the biweekly meetings, so please let me know the schedule of it.

---

<div class="post-metadata">

**Author:** ![jusong.yu](https://yyz2.discourse-cdn.com/free1/user_avatar/aiida.discourse.group/jusong.yu/32/3_2.png) [@jusong.yu](https://aiida.discourse.group/u/jusong.yu)\
**Post date:** [February 13, 2025, 2:05pm UTC](https://aiida.discourse.group/t/training-an-llm-to-generate-queries-from-natural-language-prompts-gsoc-2025/540/3 "2025-02-13T14:05:45Z")

</div>

Redirect to [Google Summer of Code (GSoC) 2025](https://aiida.discourse.group/t/google-summer-of-code-gsoc-2025/550)

---

<div class="post-metadata">

**Author:** ![jusong.yu](https://yyz2.discourse-cdn.com/free1/user_avatar/aiida.discourse.group/jusong.yu/32/3_2.png) [@jusong.yu](https://aiida.discourse.group/u/jusong.yu)\
**Post date:** [February 13, 2025, 2:05pm UTC](https://aiida.discourse.group/t/training-an-llm-to-generate-queries-from-natural-language-prompts-gsoc-2025/540/4 "2025-02-13T14:05:49Z")

</div>