______                              
(  __  \              _              
| |__)  ) ___   ___ _| |_ ___  ____  
|  __  ( / _ \ /___|_   _) _ \|  _ \ 
| |__)  ) |_| |___ | | || |_| | | | |
|______/ \___/(___/   \__)___/|_| |_|

The following experiments are being developed to explore how AI can help communities explore local data and advocate for change.

Experiment 9: Complete re-write of the API and LLM pipeline. Now using a RAG pipeline for document analysis, a SQL generation pipeline for structure data queries, and a hybrid approach for mixed queries. Refactored infrastructure now ingests documents from email and shared google folder to allow for more up to date information. New interface focused on testing the new API. Developed by a BU Spark team Fall '25.

Experiment 8: Mobile-first chat interface integrating dynamic conversation, spatial awareness, and budget and policy data along with previous data sources. Includes a separate map exploration for viewing trends. Built in collaboration with the PIT-NE 2025 Summer Impact Technology Fellowship.

Experiment 7: Spatial dashboard with generated responses integrating structured and sentiment data.

Experiment 6: Spatial dashboard with generated responses that based on selected regions and tuned for conveying community sentiment with quotes when possible.

Experiment 5: Spatial dashboard with generated caption and dialog to explore multiple factors in experiences of safety.

Experiment 4: Data dashboard with public safety and 311 livability data.

Experiment 3: Public safety data dashboard and chat bot built on community meeting transcripts.

Experiment 2: Chat interface with visualization tool to explore sample crime data.

Experiment 1: Chat interface to explore sample crime data.