Talk with Your Data: Application Development in the Era of LLM

LLMs are revolutionizing industries, but their real-world applications are still unfolding. As a researcher and developer, I'm at the forefront of this, creating an experimental cloud data analyst leveraging LLM technology. Our tool transform business queries into SQL or vector-based searches, extracting insights from our data lake, which houses Perth Crime Data, ABS 2021 Census Data, and school information. Results are presented visually or answered directly. Our demo at map.findperth.com showcases this. It features a GPT-4 backed analyst generating map reports and a chat feature for interactive queries about Perth Crime data. While promising, our prototype isn't market-ready. We're keen to share our development challenges and insights, aiming to spark discussions about the future of cloud application development.

About Pascal

Pascal Sun is a PhD Candidate at the University of Western Australia's NLP-TLP Group. His research focuses on spatiotemporal knowledge graph question answering. Prior to this, Pascal spent five years as a full-stack developer, where he successfully developed several products for global markets. Seeking greater challenges and recognizing the imminent impact of AI, he commenced his PhD journey in late 2022. Pascal is passionate about harnessing AI to simplify everyday life. In the era of generative AI, Pascal is committed to exploring how generative AI and Large Language Models (LLMs) can be employed to unlock the potential within our data, bridging the gap between complex datasets and everyday individuals.

Pascal's LinkedIn Profile

https://www.linkedin.com/in/pascalsun23/