This is a submission for the DEV Weekend Challenge: Community
The Community
I built HackFinder for the hackathon community — students, indie developers, and small teams who regularly search for hackathons to participate in.
I actively take part in hackathons myself and often help friends find events. One recurring frustration was how scattered everything is:
- Listings are spread across multiple platforms
- Filtering by location and domain is tedious
- Comparing online vs in-person events takes manual effort
- Beginners don’t know where to start
HackFinder was built to reduce that friction.
What I Built
HackFinder is a lightweight browser application delivered as a Chrome extension (popup). It organizes hackathons into a fast, searchable interface so users can find Nearby, Online, Far, and Other events in seconds.
Key features
- Domain filters (AI/ML, Web3, HealthTech, FinTech, GameDev, etc.)
- Geolocation + reverse-geocoding fallback (OpenStreetMap Nominatim)
- Distance-based categorization
- Nearby ≤ 200km
- Far ≤ 2000km
- 2-hour local cache using
chrome.storage.local - Preference for official organizer links (resolving MLH event pages when possible)
- Lightweight popup UI for quick access
The goal was simplicity, speed, and usefulness.
Demo
Short screen recording (60s):
https://drive.google.com/file/d/1i4shJug1irHr8CR1t1hq6lj02SD_jaoL/view?usp=drive_link
Code
GitHub Repository:
https://github.com/morty649/HackFinder-extension/
Note: The current submission build uses MLH and official APIs only. Support for additional listing sources such as Devpost was prototyped but disabled to respect platform terms and ensure compliance.
How I Built It
The Motivation
This project started from a personal problem. I was manually browsing multiple sites to find hackathons, comparing locations, checking domains, and filtering online vs in-person events. It felt inefficient.
HackFinder was my attempt to streamline that workflow into a single click.
Tech Stack
- Vanilla JavaScript
- Chrome Extension Manifest V3
- Popup-based UI architecture
- MLH event parsing & normalization
- OpenStreetMap Nominatim for geocoding
- Haversine formula for distance calculation
-
chrome.storage.localfor 2-hour caching
AI-Assisted Development
To build efficiently within a weekend, I leveraged AI coding assistants including OpenAI Codex and Claude Code.
I used them to:
- Refactor parsing logic
- Improve location fallback robustness
- Strengthen categorization flow
- Debug edge cases in HTML parsing
- Optimize caching structure
Rather than generating everything blindly, I used these tools as collaborative coding partners — iterating on architecture, validating logic, and refining user experience.
Engineering Decisions
- Focused the submission on MLH and official API-friendly sources to ensure responsible data handling.
- Disabled additional listing connectors in the submission build for compliance.
- Preferred resolving MLH event pages to official organizer links.
- Kept the UI lightweight and framework-free for fast popup rendering.
- Designed clear categorization to make discovery intuitive.
Why This Fits the Challenge
HackFinder directly serves a community I belong to — hackathon participants.
It:
- Reduces friction in event discovery
- Saves time
- Makes filtering intuitive
- Helps beginners navigate opportunities
Built within a weekend, it demonstrates practical execution, thoughtful design, and a real understanding of the community’s needs.
Feedback, issues, and PRs welcome:
https://github.com/morty649/HackFinder-extension/
Top comments (0)