CityChat

• 

  Venue Card

• 

  Search Results View

• 

  Chat View

Inspiration

CityChat was born from a decade in the restaurant industry, where I saw firsthand the gap between diners seeking unique experiences and local restaurants trying to reach them. Big platforms like Facebook and Google made visibility a pay-to-play game, leaving heartfelt specials and community events buried under algorithms. I knew there had to be a better way to connect people with their local food scene.

My background in community planning reinforced this vision, teaching me the importance of local engagement and the role small businesses play in community identity. It helped shape the idea of an app that doesn't just serve diners, but uplifts neighborhoods.

As a Solutions Architect, I gained the technical and strategic skills to bring that vision to life—blending tech, user experience, and data to drive engagement.

CityChat is the result: a platform designed to connect diners with local specials and events, strengthen community ties, and help small restaurants thrive.

What it Does

CityChat cuts through the digital chaos, pairing your location with venue information to deliver real-time, recurring specials and events happening around you today. Tired of endlessly scrolling through social media platforms like Facebook and Instagram, only to feel hungry and socially discouraged when searching for relevant food specials or events like Happy Hour, Trivia, Karaoke, Taco Tuesday, or Wing Wednesday? Whether you're traveling or staying in town, you want to know what's happening around you right now. CityChat is a community-based platform that delivers straightforward information about deals and events, with no unnecessary fluff, just the essentials you need to make the most of your local experience.

How We Built It

Building CityChat was a journey of relentless dedication. We embarked on tons of research, meticulously exploring the most viable technologies to create a truly scalable product. This wasn't about quick fixes; it was about laying a robust foundation. We went through countless iterations, testing what features resonated and what simply didn't work, refining the user experience with every discovery.

Challenges We Ran Into

The path to CityChat was paved with challenges:

• Scaling back features for an MVP: My vision often outpaced initial capacity.
• AI hallucinations: The guiding AI sometimes produced nonsensical or incorrect code.
• Time to work on it: A precious commodity as a passion project alongside a demanding career.
• Learning all the tools: Choosing the right ones from the vast landscape of development tools.
• Finding testers: A constant hurdle.
• Understanding the competition: Crucial for carving our niche.
• Building a usable UI: Ensuring the app was not clunky but very usable and UI-friendly.
• Supabase challenges: A significant learning curve given its dev-forward nature, a stark contrast to my Salesforce background.
• AI memory issues: The AI forgetting or overwriting existing, working code or core data, leading to frustrating rollbacks.
• Cost of scalability: Understanding the true financial implications.
• Time constraints: Preventing complex features like OCR and LLMs in the initial phase. In-app notification and automated validation model.

Accomplishments that we're proud of

We are immensely proud to have a semi-working product – a tangible tool that I, and hopefully others, can use. Even if CityChat doesn't expand beyond New Hampshire, the thought that it can help my local community, preventing another great restaurant from closing its doors, is a profound accomplishment. This is a passion project, and it will continue to be worked on, even if its path to market is slow. Each person we can bring in the door of a local eatery, each new connection fostered, means more social capital for our communities.

What We Learned

This journey has been a masterclass in a multitude of new tools and technologies. I've learned firsthand how incredibly arduous it is to build a product from the ground up, even with the guidance of AI. More importantly, I've honed my prompting skills, understanding how to better communicate with AI to achieve desired outcomes.

What's Next for CityChat

Next for CityChat is the integration of a seamless, interactive map. Beyond that, we envision a "recency model" for ad-hoc advertisements, designed to undercut the traditional pay-to-play models of Google and Facebook. Imagine this: if someone is actively using the CityChat app, they're likely hungry and looking for a dining experience. If a restaurant can, with a single click, promote a real-time special to users within a 10-mile radius of their location, that's hyper-local, in-the-moment advertising that truly connects. This is about delivering exactly what a hungry customer is looking for, precisely when they're looking for it.

Built With

• javascript
• mapbox
• netlify
• openai
• plpgsql
• supabase
• typescript