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:
An ML research agent that runs the whole loop on its own; and never hides a bad result.
A robust AI-generated image detector that generalizes across unseen generators and real-world transformations—not just a single dataset.
Built in Colab/VS Code, our PyTorch ResNet50 detects AI images after noise, JPEG, blur, crop and resize. We used torchvision, scikit-learn and pandas, trained on CIFAKE and SID_Set; no external API.
Aegis sees people and boats in the water, then dispatches help, reroutes rescue, or calls a human.
AI image detection that survives the internet
A real-time risk analysis and early decision engine powered by Shannon entropy, deployed on Google Cloud for fast, data-driven outcomes.
glhf
An LLM agent that diagnoses, changes, retrains and measures a recommendation model in a loop: 0.6016 → 0.6056 on KuaiRand-Pure, no human intervention, and a harness built to check its work.
Just tell us what you need. We'll figure out what you're looking for.
ChurnGuard — Predict churn. Protect customers.
AI can fake reality. We built a detector that doesn’t just work in the lab—it stays accurate when AI images are edited, compressed, and shared online.
A Recommended Systems ML research agent that diagnoses instead of guessing results from a training loop. It reports the baseline, forms its own hypotheses, and records why its ideas fail.
Needle in a haystack, solved by conversation
A CLIP-based AI-image detector, rigorously stress-tested for robustness to real-world compression, blur, noise and cross-generator shift — failure modes measured, not hidden!
AI shopping agent that turns vague preference into the right product with high efficiency
"An autonomous agent that does the work of a machine learning engineer: given a dataset and a metric, it writes code, trains models, reflects on results, and iterates—all on its own."
A conversational product recommendation system which is like a shopping assistant. Understands user’s intent, keeps track of session state, and asks follow-up questions before recommending products.
Detect scams before they detect you
RippleGraph uses Graph Neural Networks to predict how supply chain shocks cascade through connected companies — turning 'what if X breaks' into a quantified risk score in under 100ms.
Your personal shopping buddy that listens to you.
A detector that survives real-world image damage and generalises across AI generators. One extra training source took cross-generator accuracy from 49% to 97%, at a cost of 0.5 points.
Roshan Grid uses AI agents to forecast local energy needs and coordinate national power dispatch, helping Pakistan use the cheapest available energy while improving grid efficiency and reliability.
"BreathPrint turns a 30s phone recording of a cough, breath & vowel into instant, explainable respiratory risk triage—running 100% offline to empower frontline health workers."
because seeing shouldn't mean believing