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
Independent builder · AI products, hackathons, comedy, and human behavior
Practice high-stakes meetings before they happen. Present your case to an AI panel, answer tough questions, and watch how their private deliberation turns into a decision.
Transforming raw, long-form comedy footage into high-engagement social media content by leveraging the multimodal reasoning of Gemini 3 Flash.
Production Pilot reviews video submissions before they reach the producer, catches technical issues, and tells creators exactly what to fix before submitting.
Stage Ready uses OpenCV to inspect raw video, fix framing and presentation issues automatically, and tell creators when a reshoot is needed.
A simulated Alexa+ experience that detects rising scam risk in conversation and shifts naturally from clarification to protective assistance.
Clarity finds the one vague claim worth asking about, then uses a short adaptive phone call to turn it into a clearer fact, backed by the candidate’s own words.
Audition Agent turns an AI-generated character into a human casting workflow, creating auditions, organizing applicants, and handling follow-up while keeping casting decisions human.
Save Peps is a one-tap 3D puzzle game where you choose the right object to reunite two lovable Peps across 12 increasingly wild worlds.
Payoff helps creators design what an audience should feel, then uses real audience reactions to discover whether the story actually got them there.
FaceState turns one selfie into personalized All-Nighter, Time Jump, and Character transformations, using YouCam skin analysis to make each result feel specific to you.
Replaceable is a living map of how AI is changing work, showing which jobs may be affected sooner, why, and how those assessments change as new evidence appears.
Helps judges, recruiters, and program reviewers evaluate submissions faster with structured, evidence-backed analysis while keeping scores and final decisions human.
Beatboard turns a comedy idea into a shoot-ready storyboard with timed beats, generated art, voiced dialogue, direction, and dependency-aware revisions in minutes.
MotionKey lets you teach your phone custom hand gestures, train a tiny personalized model on Arm64, and use those gestures to control the phone offline.
Turns a technical builder’s real experience into evidence-backed, editable applications tailored to hackathons, fellowships, accelerators, and other selective programs.
It maps a project's files to the claims they support in DataHub, then an AI agent reads that map and reports which claims the evidence proves. When it can't tell, it says so instead of guessing.
Give it a project description and its demo video. It watches the video and makes a short narrated clip showing which claims the demo actually proves, which it skips, and which it contradicts.
Send in a project write-up and get back a report on which of its claims the README and docs actually back up, with quotes as proof. Costs 5 cents per check, paid automatically with x402.
Trajectory turns long ChatGPT conversations into a clear arc, useful findings, and evidence you can trace back to the exact messages.
Save Pip in one tap. Play three tiny rescue puzzles, see instant results, retry freely, and chase a perfect run.
A Slack agent that turns high-stakes provider updates into one decision card. Reviewers ask questions, request a real verification call, and approve changes before families rely on them.
Seven Qwen agents investigate a family's complaint about a Medicaid therapist listing, verify it using real sources and a phone call, then publish only evidence they can defend.
Every opportunity becomes a case. Mission Control automatically resolves routine intake and escalates only the cases where human judgment creates greater value.
Pulse asks six rival AI models the same builder questions about any model, then publishes their competing takes as one clean page on Aurora PostgreSQL and Vercel.
