
Developed an end-to-end information extraction notebook for the Esri/arcgis-python-api repository, focusing on automating the analysis of Cheshire Fire Incident Reports. The solution leveraged Python and Jupyter Notebook to integrate the Mistral large language model with the arcgis.learn API, enabling extraction of key entities such as addresses, dates, and incident types from real-world text data. The workflow encompassed data preparation, model training, evaluation, and inference, demonstrating practical application of natural language processing and machine learning within a GIS context. This work addressed the need for automated, accurate data capture to support faster incident analysis in public safety operations.
January 2025: Delivered an end-to-end Information Extraction Notebook for Cheshire Fire Incident Reports leveraging the Mistral LLM and the arcgis.learn API. The notebook covers data preparation, model training, evaluation, and inference, focusing on extracting addresses, dates, and incident types from real-world text. This work demonstrates applying LLMs to practical public-safety data analysis tasks and showcases the value of NLP-driven automation within the Esri/arcgis-python-api ecosystem.
January 2025: Delivered an end-to-end Information Extraction Notebook for Cheshire Fire Incident Reports leveraging the Mistral LLM and the arcgis.learn API. The notebook covers data preparation, model training, evaluation, and inference, focusing on extracting addresses, dates, and incident types from real-world text. This work demonstrates applying LLMs to practical public-safety data analysis tasks and showcases the value of NLP-driven automation within the Esri/arcgis-python-api ecosystem.

Overview of all repositories you've contributed to across your timeline