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
CaseProof is a Gemini-powered decision copilot that turns medical bills and denials into cited, prioritized, human-reviewed actions—helping patient advocates serve more people safely.
KeepMe makes AI virtual try-on trustworthy. Shoppers control what can change, detect unwanted changes to their appearance, repair visual drift, and receive proof before approving the result.
Offline voice intelligence for emergency teams—turning noisy radio reports into reviewed, prioritized incidents on local Arm hardware, even when connectivity fails.
When data consent changes, RevokeGraph finds every affected model, orchestrates an approved replacement, and proves the withdrawn data no longer reaches production.
The safety layer for generative recall media—turning approved notices into verified voice and visual packages while blocking factual drift and preserving evidence in Backblaze B2.
RedressCI is the remediation platform that turns reported AI failures into privacy-safe, evidence-backed regression tests, verifies fixes, and prevents them from returning in production.
The evidence operations platform for IEP service delivery—turning fragmented records into an auditable ledger, grounded review signals, and human-approved action packets.
Open one handmade surprise. Send one back. Parcel turns Reddit into a daily gift exchange between strangers.
Precedent turns Slack conversations into human-confirmed decision memory, so people and agents know what was decided, why it changed, and what is still true.
Agents, robots, and humans break at the handoffs between them. SeamProof tests those seams against the real run trace, gates the release, and its Seam Analyst agent recommends the fix.
Civic Pulse is a low-cost emergency routing desk for nonprofits. It turns incoming community needs and live provider capacity into explainable referrals so people are sent to help to the right place.
Most flagged vulnerabilities can't actually be exploited. PathGuard walks GitLab Orbit's call graph backward from each one to your entry points, and posts only the handful that are truly reachable.
Turn live Splunk data into an evidence-backed incident timeline, AI‑ranked root causes, and an auto‑drafted postmortem — agentic incident command that cuts triage from an hour to seconds.
The safety layer for the AI nobody's watching — Sentinel catches a chatbot's confidently-wrong answers before they reach a vulnerable person, every claim checked, every decision auditable.
Most mod tools remember users. ModCase remembers decisions — it captures your team's calls automatically and shows you the precedent before you moderate.
ToxFlow is the smart-money radar for Polymarket. It scans live markets in real time and shows you where informed traders are moving — institutional orderflow intelligence, before the crowd.
When care teams or AI recommend follow-up, LoopGuard makes sure someone owns it, tracks it, and closes the loop.
StoryBridge helps Team USA fans discover Paralympic stories connected to the Olympic sports they already follow.
A Logitech Actions SDK plugin that turns your MX Console into a physical consent firewall — blocking AI agents from executing risky actions without a hardware-signed human approval.
Institutional-grade orderflow toxicity detection for Polymarket using VPIN — the same technique that detected the 2010 Flash Crash, now adapted for prediction markets.
Magnitude-gated AI agent authorization — tokens scoped to scope x reversibility, with mid-action self-invalidation when an agent discovers more items than authorized.
AI conversation debugger that detects when discussions loop, visualizes repetition patterns, diagnoses hidden assumptions, and generates the one question that breaks the cycle.
VPIN-based orderflow toxicity engine for Polymarket. Detects informed trader activity using volume-synchronized analysis, confirms with Synthesis market data, and generates actionable trade signals.
diffRact - As AI-generated code increases, the review process needs be smarter too. The agents analyze code issues, trace affected code, flag risk with recommendations, and estimate wasted CI compute.
