Devpost
Participate in our public hackathons
Devpost for Teams
Access your company's private hackathons
Grow your developer ecosystem and promote your platform
Drive innovation, collaboration, and retention within your organization
By use case
Blog
Insights into hackathon planning and participation
Customer stories
Inspiration from peers and other industry leaders
Planning guides
Best practices for planning online and in-person hackathons
Webinars & events
Upcoming events and on-demand recordings
Help desk
Common questions and support documentation
Sort by:
Check your ebook before you upload it: every EPUB problem explained in plain English, the common ones repaired in one click, and the file never leaves your computer.
Cardie runs the FDA’s ORd-CiPA heart-cell model on real Li et al. 2017 patch-clamp drug data to predict arrhythmia risk - simulated, stored, and compared live in Tiger Data (TimescaleDB).
We trick AI agents that pay invoices into paying fraudsters, then fix them and prove the fix holds on attacks they never saw.
Know Your Glucose. Live Wiser: Turn your CGM data and meal photos into your personal AI glucose assistant.
An AI-powered civic reporting platform that helps citizens report local problems, track resolutions in real time, and build smarter, safer communities through insights and transparent governance.
Webcam-based rehab tracking for knee and hip patients: real-time pose estimation measures exercises and range of motion.
Nerevana is a B2B marketplace connecting small Kenyan produce vendors to institutional bulk buyers like schools, turning wasted harvests into stable revenue and fresher supply.
AI turns lecture slides into gamified 3D worlds for neurodivergent learners. Earn MOE on Solana by learning, send it peer-to-peer, and buy and sell personalised worlds from other students.
A project helping neurodivergent kids learn.
The credit notebook for small shops: one sentence in, a clear balance and a polite WhatsApp reminder out.
Helping humanitarian teams respond to displacement in Sudan by predicting food shortages, comparing ways to distribute aid, and showing how delays and transport constraints affect people in need.
The sun has receipts. AI checks solar claims against weather; smart contracts block duplicate asset-hours. Portable evidence makes each claim open to inspection.
Your life, your memories, your evolving digital self.
Horizon uses a calibration model to correct how overconfident Polymarket gets near close, and an AI agent that finds the mispriced bets, sizes them and fills the order for you.
Ctrl+Z for the physical worldFor someone living with dementia or ADHD, losing their glasses or wallet 10 times a day costs them their independence. Rewind quietly remembers where every item ended up
A safe, self-improving AI finance team: it applies payments, chases overdue invoices, pays vendors, blocks fraud and catches its own mistakes. It accumulates experience and takes it to new companies.
Check local shop stock, prices and trends before you leave home. A simple stock tracker for students and small shopkeepers to mitigate wasted trips to grocery shop for those who have just started.
SikaGuard: An AI-powered security firewall that protects users from deceptive financial transactions by verifying evidence before autonomous agents spend their money.
Calls whether Britain's grid will be short or long an hour ahead, from public data only, and puts a price on what that call is worth to a battery.
Ever felt stuck in Lenton? Wanted to escape from the area and it's lentonians with an awesome car that can jump? This just might be the game for you!
Commit Explorer turns any public GitHub repo’s history into a Windows XP-style graph: size shows how much changed, colour shows what kind, and a time-lapse replays how the project grew.
Let's make our mistakes in the model - so that when disaster strikes, we are prepared. This model acts as a tool for guidance in the handling of potential outbreaks of disease; keeping you safe :]
IF – INTELLIGENCE FACTORY CODING AGENT - From Ideas to Intelligent Applications
Forecasting what Wikipedia will read next hour so caches keep the right articles: 9.4% fewer requests hit the servers than with the standard method, winning all 72 test hours.
