Inspiration
The development of Hiraku came from a multitude of factors: hearing about the recent October 1 shift, the remarkable but unseen stories of SNAP recipients not getting enough closure on their benefits; all of it drove our intent to create an accessible and comprehensible experience to navigating SNAP.
What it does
Hiraku helps New York City residents understand changes to SNAP and find official next steps when benefits stop or a household’s benefits are reduced. It presents guidance in a clear, audience-focused format and links people to the appropriate official resources. A separate Explore demo shows how grocery discovery might work, but it uses fictional stores and illustrative distances; it does not find verified retailers or determine SNAP eligibility.
How we built it
We built an Info-first web experience, with a separate map-and-list demo for grocery discovery. We tailored the content and navigation to common SNAP questions, and kept the guidance focused on official handoffs rather than making eligibility or benefit promises. The Explore demo uses clearly labeled sample data. We also developed a retailer API as a separate workstream, while leaving its connection to the frontend for later.
Challenges we ran into
The hardest part was making the guidance useful without implying that Hiraku can determine someone’s eligibility or resolve their case. We also had to be precise about what the grocery map represents: sample locations and distances are not verified store information, SNAP authorization, or route guidance. another challenge we would say was keeping the frontend and backend scopes distinct and making the remaining data and integration limits clear.
Accomplishments that we're proud of
Being able to look at a project, constantly scrutinize it, and weigh in all the different factors that make a hackathon
What we learned
Hiraku taught us that writing code isn't always the hardest part of software engineering. Knowing what to build, why to build it, and how to validate it can be much harder. Modern AI-assisted development tools make implementation increasingly accessible, but they don't eliminate the need for sound engineering judgment.
Through this experience, we learned several valuable lessons:
Problem validation comes before implementation. Researching the problem helped us identify gaps between our assumptions and the evidence available.
Technical excellence includes knowing what not to build. Adding sophisticated features means little if they don't meaningfully improve the user experience.
Good product design requires understanding tradeoffs. A nearby grocery store isn't necessarily affordable, and an affordable destination isn't necessarily accessible.
Planning and architecture are force multipliers. Working with structured engineering workflows helped us make more deliberate decisions and identify weaknesses earlier.
The biggest lesson we took away was that software engineering isn't just about producing a working application. It's about translating messy, real-world problems into thoughtful and trustworthy solutions.
What's next for Hiraku
Improve accessibility and language access. Test with screen readers, mobile layouts, and people who prefer languages other than English.
Measure where Hiraku can improve in aiding SNAP recipients. Track whether users can find the right official next step and understand the limits of the information without collecting sensitive case details.
Built With
- next.js
- vercel
