Check the 'Available Agendas & Minutes' tab/link above to see them all. But basically, most 2024 and 2025 Framingham city government meeting agendas minutes that have been posted (there can be significant delays for minutes, and some minutes files don't seem to get posted at all) from the:

• City Council (including subcommittees)
• Bicycle, Pedestrian & Trails Committee
• Board of Health
• Board of Library Trustees
• Board of License Commissioners
• Charter Review Committee
• Council on Aging
• Disability Commission
• Framingham Sustainability Committee
• Human Relations Commission
• Planning Board
• Police Advisory Committee
• Sustainability Committee
• Zoning Board of Appeals
• and a few Community Preservation Committee meetings

You should be able to ask things like

• Did Getchell trail projects get CPA funding? If so, how much? Note: Asking for all the projects approved may only get you a partial list, because this app can only process a limited amount of text at once. However, the answer should include a link to minutes with the information.
• What has been said about Nobscot and the MBTA Communities Law?
• Tell me about drug overdoses in Framingham.

However, be advised that the generative AI may not always fetch the most recent data since it's looking for the most relevant data.

A format like 'What happened at the Bicycle, Pedestrian, and Trails Committee meeting in February 2025?' should work, but you need to give the exact board name. Times like 'last month' -- for example, 'What happened at City Council meetings last month?' should work if minutes have been posted. Search the Available Agendas & Minutes tab above to see what's here. But AI can be unpredictable! Also note that questions with a time component take a lot longer to fetch.

You can do a conventional text search of agendas and minutes at https://apps.machlis.com/shiny/framingham_meetings

Data come from PDF files of minutes posted on the Framingham goverment website, which are then converted into text.

It first analyzes your question and 'translates' it into a series of numbers called embeddings, and then checks the documents to find excerpts with embeddings that are most similar to your question. It also tries to use some traditional search similarity techniques, too. Then, your question and those releative text chunk are sent to a Large Language Model (same technology as behind ChatGPT) to generate an answer.

Technically Sharon Machlis, but in large part this was coded with generative AI 😅 using Google's Gemini 2.5 pro large language model with help from Posit's Shiny Assistant and occasionally Claude Sonnet 3.7 and various ChatGPT models and custom GPTs.

This is not an official city of Framingham app.

It currently uses OpenAI's GPT-4.1 (full model) released in April 2025, which excels at instruction following, has a large 1 million token context window, and is optimized for accuracy and consistency.

This app uses the R Shiny Web framework and a lot of cool R packages including shiny, shinychat, ragnar, ellmer, and reactable.

There are many efforts underway to create chatbots like this for a lot different use cases, such as software documentation and customer service. We have one at work answering questions about all our articles from Computerworld, CIO, CSO, InfoWorld, and Network World called Smart Answers . There's also another, consumer tech version for PCWorld, Macworld, and TechHive if you want to play. These applications are known in the field as RAG (Retrieval Augmented Generation).

I also created a Framingham Recyclebot app .