PitchInNYC - Volunteer Finder

Inspiration

Volunteer opportunities in NYC are often scattered across nonprofit websites and platforms such as Eventbrite and Idealist. Finding the right opportunity can take significant time and effort.
That inspired PitchInNYC, a single place to discover volunteer opportunities.
Our goal is to make discovery easier while directing users to the original source to verify details and apply.

What PitchInNYC Does

Users can search and filter opportunities by cause, borough, and availability.
They can also use our AI Matchmaker to receive recommendations based on their profile and preferences.
Users then continue to the original nonprofit website or platform to review and apply.
This keeps users in control while AI supports discovery and recommendation.

How We Built It

We used Google AI Studio to turn our requirements into an initial prototype.
We continuously tested the app and refined features through iterative prompting.
As we learned, we shifted from giving AI direct instructions to asking it to propose multiple solutions.
We evaluated those options as a team and kept final decisions with the human team members.

Challenges

AI sometimes misunderstood our prompts or generated features that were not clearly useful.
It also produced incorrect information and links, requiring repeated human testing and verification.
Our prototype is not yet connected to live volunteer APIs, which limits real-world data validation.
We also faced GitHub sync issues and had to manually export and upload our project files.

What We Learned

We learned that building with AI is a collaborative and iterative process, not just prompting.
AI helped us quickly turn an idea into something tangible, even without traditional app development experience.
Seeing the prototype early helped us refine our requirements and user experience.
We also learned that human judgment, testing, and verification remain essential.

What We Accomplished and What's Next

We completed a working prototype of Phase 1: Discover during the hackathon. Our vision is to develop PitchInNYC in four phases:

Phase 1: Discover

Phase 2: Connect & Apply

Phase 3: Plan

Phase 4: Go Mobile

Built With

chatgpt
claude
gemni
github
googleaistudio

Submitted to

Built for NYC: AI Hackathon at NYPL

Created by

I came up with the initial idea for PitchInNYC based on my own experience struggling to find volunteer opportunities in NYC. Drawing from that experience, I contributed many of the ideas that shaped our initial product requirements, including what information each opportunity should contain, how users could search and discover opportunities from multiple sources, and what would make the experience genuinely useful for someone looking to volunteer.

Using those ideas, we built the initial prototype in my Google AI Studio account, turning our early concept into something tangible that we could test and discuss as a team. We made several revisions until we reached the free token limit.

On the second day, I focused primarily on preparing our hackathon application and Devpost submission, including the elevator pitch, project story, challenges, and key learnings. Meanwhile, my teammate, Ming-Yi, focused on further improving the user experience, performing quality checks, and making additional changes to the prototype.

I also contributed to shaping the broader product direction, including our focus on discovery rather than replacing existing application platforms, responsible AI principles such as source verification, human review, and user feedback for incorrect information. I also proposed product features to help prevent scheduling conflicts and double-booking when users plan multiple volunteer commitments.

Naomi Su
Once Naomi come up wth the initial idea, I put on my curious hat and started trying to understand her pain points of what problems she was experiencing in looking for volunteering opportunities. Based on her responses, I wrote down our initial list of requirements and fed it into AI Studio to create the initial Prototype.

The LLM generated some functionalities that required some manual review. I went through each of the functionalities and evaluated their usefulness and whether they made sense at all. Based on my evaluations and discussions with Naomi, we continue to make modifications.

I also took a step back and considered the user journey and split the application into 2, one for the individuals looking into volunteering opportunities and another is the organization that would be able to post the job opportunities.

Ming-Yi Smith

Updates

Naomi Su started this project — Aug 16, 2026 01:53 PM EDT

Leave feedback in the comments!

Log in or sign up for Devpost to join the conversation.