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:
TICK TOCK, time is compute ⏱️ Baseline? Just the starting line 🏁
A dynamically dispatched, hardware-aware Transformer kernel suite that achieves 11.97x speedups peak and processes 100k tokens by exploiting NVIDIA architecture and custom Triton-Gluon fusion.
Beyond Words, Beyond Worlds.
An ML research agent that edits a typed operator graph instead of a script, making broken pipelines unrepresentable. Beat the KuaiRand-Pure baseline in 4 unattended iterations, 0 interventions.
An agentic system that finds every legal front to reduce or erase a medical bill and files the paperwork itself.
A robust AI-image detector designed to withstand real-world transformations and cross-domain shifts, with reproducible evaluation and an interactive demo.
AI images are everywhere nowadays. Our detector checks if an image is synthetic, whether it be compressed, blurred, resized or edited, so people can question what they see, not blindly trust it.
TraceGuard helps spot AI-generated images—and tells you when its own answer might be unreliable after an image is compressed, resized, or reposted.
Don’t trust the first patch. Prove the best one with parallel Nemotron repairs and reproducible execution evidence.
AI Problem Solver is a simple idea: instead of giving you another chatbot that just talks about a problem, it helps you actually figure out what to do next.
Two retrievers, one literal and one semantic, fused by how confident each is rather than blended blindly. Narrows 50,000 products to the right one in about two questions.
Scrape live customer demands business categories, create company pages, contact leads via Email, WhatsApp, or autonomous A2A Agents, converse with buyers, and generate custom proposals.
Talk to My PC is a voice interface for your computers. Call in from anywhere, ask what your machines are doing, and manage their work through a controlled set of tools.
An AI shopping copilot with fast deterministic reflexes, autonomous LLM reasoning, and a confidence controller for hard conversations.
Our project fine-tunes a pre-trained ResNet18 model to detect AI Images from real images.
search gives you ten blue shoes. needle asks the question that makes nine of them irrelevant.
A live status map for Sydney's food banks and shelters. Not just where they are, but whether they actually have anything on the shelves right now.
An offline-first shopping agent that asks better questions, preserves changing intent, and turns catalog evidence into earlier, higher-ranked product matches, and using zero runtime LLM tokens.
A lightweight AI-image detector built to survive what happens after upload: compression, cropping, reposting. The real test most detectors skip.
A verbatim-evidence shopping agent that finds the hidden target among 50,000 products in ~2.5 turns: 0.962 vs 0.107 baseline, HR@10 = 1.000, deterministic and fully offline.
A conversational shopping agent that hits 200/200 — because we found the bottleneck wasn't language.
Conversational shopping search that remembers, adapts, and recommends fast.
Offline reliability, local AI intelligence
“An autonomous AI research agent that investigates complex questions, analyzes evidence, challenges conflicting claims, and delivers trustworthy, evidence-backed insights