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:
The AI suggests answers from redacted text; Midnight proves your disclosure policy was enforced.
AI agents that deliver code through pull requests — Claude Code + Codex with worktree isolation, a command allowlist, and secret scanning. Never automatically merged.
Accessible campus navigation that adapts routes to mobility needs and live disruptions, helping users avoid stairs, steep slopes and unavailable elevators.
Your silence can still call for help. SilentSOS turns authorized CCTV into an AI-assisted early-warning system that detects distress, adds real-time context, and coordinates rapid emergency response.
Finding opportunities means wasting hours on expired listings and complex requirements. Taskmaster uses a multi-agent Gemini pipeline to research, filter, and rank student tech path options.
Prove you're old enough without revealing your birthdate to a server. ProofPass swaps a rental app's date-of-birth field for a Midnight zero-knowledge proof; only pass/fail ever touches the chain.
Most shopping assistants ask what you want. Ours works out what it doesn't know and asks that instead.
Narrow is an offline shopping agent that asks the smartest question each turn, narrowing 50,000 products to your exact match in about 2 turns, no LLM, no API cost with an 100% product hit rate.
Vumbi AI: An autonomous marketing agent that handles busywork, learns from outcomes, and only surfaces for decisions.
Shopping search that listens: our agent remembers evolving needs, handles changes of mind, and combines BM25 with semantic attribute ranking to understand shoppers beyond exact keywords.
Turning emergency situations into structured, actionable information.
RealityCheck is a practical AI-image detector that helps users assess whether an image is likely AI-generated even after common transformations.
An intelligent AI assistant designed to help users with coding, learning, research, and everyday problem solving.
Sport's easy to love and hard to organise. MatchUp finds you a local pickup game, shows who's going, and lets you claim a spot before it fills, or post your own and let the roster fill itself!
Corotum helps you keep AI agent skills consistent across every machine and AI agent. Instead of copying and fixing drift, you manage one desired state and sync it safely through Git or Corotum Cloud.
Where researchers and their AI agents triangulate real DBS research evidence together.
Nodoryx is a building-energy digital twin that detects abnormal power use, predicts overload risks, and recommends safer actions before small energy problems turn into outages.
Finds the right product in 3.3 turns with zero tokens, zero network, and no LLM. Deterministic retrieval plus one well-chosen question. 8× the organizer baseline.
An autonomous recommender-systems research agent that plans, tests, evaluates, and documents evidence-driven model improvements.
Your domain already replaces Bitly, Linktree & Google Workspace ($56/mo). LaunchKit makes it easy: AI naming, branded links, email, DNS, security & brand protection — no agency needed.
Every Agent Run is guilty until proven safe.
Team CAPybara knows when someone's capping. So with our skillset, is it any surprise we went for Track 5? With our solution, you can clock what's real, because getting fooled by an AI? Now thats an L.
A robust AI-generated image detector designed to generalise across unseen generators and remain reliable under real-world image transformations.
A Transformer bottlenecks differently at every input shape. We time the candidates at runtime and cache the winner per shape: ~10x geomean speedup, 13/13 correct, on a laptop GPU.
