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
StorySpark AI is an AI-powered platform that turns one prompt into stories, songs, storyboards, voice scripts, and educational content for children.
KAAAGLUMN Basilica is a tutelary memory agent that persists observation across sessions and cross-checks other AI conversations against actual repo state. Memory as guardianship, not just recall.
A granular agent that helps you keep your memory in a way a computer does
The memory system that forgets. So what matters stands out.
A decentralized multi-agent AI framework that turns controversial football match incidents into transparent, unbiased, and verifiable decisions in real time.
ForgeOS is an AI operating system that transforms isolated LLMs into persistent organizations with memory, mission execution, and collaborative reasoning, powered by Qwen on Alibaba Cloud ECS.
An agent that catches statistical errors in a paper before submission. Qwen reads the numbers; exact math makes every verdict, so no result can be hallucinated.
Qwen-powered memory that keeps confirmed project judgments from being silently overwritten.
Four AI agents argue before they answer. Memory, logic, and ethics debate in real time, a judge agent decides, and persistent memory on Alibaba Cloud OSS makes every session smarter.
Zero-LLM procedural memory for coding agents: concept graph with tick-decay + file:line provenance, plus Qwen Cloud for grounded answers. Remembers how past bugs got fixed.
A memory that stops AI coding agents from re-introducing bugs they already fixed — confidence grounded in real test outcomes, not the model's opinion.
An autopilot agent that reconciles Paystack payments against invoices with Qwen, auto-closes confident matches, and escalates the rest to Telegram — never guessing when it isn't sure.
A Qwen-powered Autopilot Agent that turns messy B2B customer inquiries into safe, human-approved quote PDFs with Alibaba Cloud deployment.
Autonomous Site Reliability Engineering Agent
**Beubble 2.0 is a multi-agent AI platform that enables organizations to use collaborative AI agents capable of executing tasks, connecting systems, and maintaining persistent intelligence.
An explainable Qwen-powered caseworker that turns ambiguous life stories into verified social-rights assessments and human-approved action plans.
"A multi-agent hiring pipeline featuring double-blind screening, fairness auditing, and autonomous interview scheduling."
HealthClick is a multimodal AI health companion. Chat, speak, or snap a photo of a symptom; it analyzes the condition and instantly drops the nearest hospital’s name, phone, and directions into chat.
Qwen triages real county food inspections, drafts the follow-up, and a human approves every action. Backtested on the county's own history: URGENT-tiered facilities failed next 3x as often.
AI helps shop under a human constitution — no invented products, no rule changes.
Qwen supervises the traffic lights, the edge runs them. Qwen reasons over incidents and proposes policy, a human approves. When traffic stops looking normal, Qwen makes it flow again.
From a headline to the official record—Qwen verifies what public evidence can and cannot support.
Project memory that sticks — across sessions, tools, and decisions.
Autonomous AEM content agent that finds stale pages, refreshes them with Qwen AI, and writes back to JCR — with human approval before anything goes live.
289 – 312 of 729
