citymeetings.nyc

What is citymeetings.nyc? Who made it?

Check out the about page to learn about the project.

Where do these meetings come from?

They're downloaded from Legistar, which the city council uses to manage their meetings and legislative process.

Can you cover community boards/NYC planning/other local government meetings?

There are so many! Cast a vote by sending me an email at vikram@citymeetings.nyc with what you want and why.

I'm figuring out how to do this scalably. This includes improving my AI-powered internal tools and seeking funding through consulting and sponsorships.

Do you take feature requests?

Yes! I want to hear from you.

I may not fulfill every request, but I'll consider and reply to them all. Please send your request to me at vikram@citymeetings.nyc.

How do you determine what a "chapter" is in a meeting?

I want chapters to be granular. Viewers should be able to browse/skim and find information they care about.

For example, instead of "an exchange between council member X and agency Y", which might span 30 minutes, I strive to have chapters that cover specific exchanges/questions under 5 minutes.

This doesn't always apply/make sense, but it's the rule-of-thumb I use.

How do you create all these chapters? Do you watch all the meetings and do them manually?

I use a combination of AI tools I've built and human review (me!) to determine chapter boundaries and write titles/descriptions.

I go into details on how this all works on my blog, where I've published annotated slides for a talk I delivered at NYC School of Data.

I see a bunch of transcription errors. What's up with that?

I use Deepgram for transcription and diarization (a fancy word for "identifying when different speakers are speaking"). They have the best quality-to-price ratio I found, but it's better-trained for other types of audio/subject matter.

I'm a solo operation so I can't manually review and fix mistranscriptions. Further, large language models (LLMs) are quite good at picking up on transcription errors using context clues, so my tools don't suffer from them.

I have plans to try and use LLMs to fix transcription errors, but that is a large body of work and I'm prioritizing other things for now.

If you see an egregious transcription error (your name is totally wrong, or a mistranscription is offensive/borders on it), email me at vikram@citymeetings.nyc and I'll fix it.

I see errors or inaccuracies in chapter summaries. How do I report them?

Email me at vikram@citymeetings.nyc and read below for my policies, which I'm still figuring out and are subject to change.

My intent with chapter summaries is to represent things people have stated on the record at government meetings in a way that is easy to skim.

If a chapter summary is inaccurate because it summarizes an inaccurate thing that something someone said, I won't correct the inaccuracy. It's on the record.

I don't do or show fact checks today. I might in the future.

I attribute claims made in summaries to speakers in meetings. These claims are not citymeetings.nyc's own claims. I admit that the summaries and UI on citymeetings.nyc do not always make this clear and I'm working on it.

If I have inaccurately represented something someone said, I'm keen to fix it. This includes errors of omission. There's a tension between summary skimmability and how faithful they are to what someone actually said. I want to strike the right balance.

It's common to see misspellings of names, agencies, or specific details because of mistranscriptions. I'm keen to fix these in my chapter titles and summaries, especially if it is your name!

If you've been in a position where you've implemented and enforced similar editorial policies, I'm new to this and would greatly appreciate your advice. Email me at vikram@citymeetings.nyc.

Can you work with my organization to run workshops on AI, or to build similar AI-powered data extraction and query tools on our private data/other public datasets?

Yes.

I've been an engineer/tech lead/executive for ~20 years and have been a CTO-for-hire for several years.

I have plenty of experience working with organizations (startups in particular) to ship useful software/build teams, with glowing reviews from clients who I'm happy to put you in touch with.

I am also available for presentations, workshops, and advisory work for leaders/teams who want to use LLMs more effectively or want to build tools/products/services with them.

My focus is on deeply practical takeaways that you can use. Watch my talk at NYC School of Data 2024 (March 23rd, 2024) for what this looks like.

Email me at vikram@citymeetings.nyc.

My organization benefits greatly from this work. I'd like to support it/I'd like for you to invest in features I want and meetings I'm interested in. How can I do that?

I'm glad your organization finds this work valuable! You can sponsor my work.

Funding from your sponsorship will help me scale citymeetings.nyc and give you a say in the roadmap around features I build and meetings I cover.

Email me at vikram@citymeetings.nyc to discuss sponsorships.

I'm an individual and I'd like to support this work. How can I do that?

Aw, shucks. Thanks!

I'd love for you to spread the word about citymeetings.nyc, sign up for the newsletter, and request features.

If you want to support this work financially, I'm looking into ways to do that with a package of features useful for interested citizens and professionals who use this work.

Email me at vikram@citymeetings.nyc to let me know you're interested in supporting this work as an individual and to discuss features that'd help you.