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:
bbq sauce has designed a shopping copilot to find the perfect product for you in a Top 10 list in roughly 3 turns, with no LLM API, fully offline, with every design decision backed by data.
TrueFrame detects AI-generated images by fusing CLIP visual embeddings with FFT frequency signals, improving robustness to compression, blur, resizing, noise, colour shifts, and cropping.
Mercury helps you find the correct product through a conversation. It asks useful questions, remembers what matters and adapts when your preferences change without making you repeat everything again.
An end-to-end short-video recommendation pipeline that combines LLM embeddings with tabular features using SENet-DeepFM architecture and dynamic hybrid loss.
RobustLens estimates whether an image is AI-generated after layers of post-processing using calibrated multi-view scoring, transformation-chain testing, and safe abstention when evidence degrades.
Robust Detection of AI-Generated Images Under Real-World Transformations.
Dual-stream AI-image detector: CLIP reads what it shows, frequency analysis reads how it was made. Trained to survive compression/cropping, and honestly tested on generators it's never seen.
A conversational e-commerce search agent, touching lives, making shopping easier.
AI Image & Video Detection Built For The Wild
Shadeway maps the coolest path through a city, helping people avoid heat and walk safer in a warming world.
An autonomous agent that does ML research: hypothesise, code, run, measure, repeat. It beat the official KuaiRand recommender baseline in 30 self-directed iterations with zero human intervention.
A transformation-aware AI image detector that combines semantic and forensic evidence to remain reliable after compression, resizing, and real-world image transformations.
An autonomous ML research agent for KuaiRand-Pure that proves its improvements. A paired-seed bootstrap gate rejected five hypotheses and accepted one — numpy only, zero GPU, zero LLM tokens.
Zero-leakage topological Graph ML platform for crypto forensics, engineered to detect laundering syndicates where standard models collapse under temporal drift.
Reliability Enterprise Lifecycle Operating System: Autonomous multi-agent pipeline integrity governance combining deterministic engineering physics solvers with Google Gemini 3.5 Flash.
An ML research agent that runs the whole loop on its own; and never hides a bad result.
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
"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."
Sing along to any vocal and get matched to its real notes, built with vocal alignment and a vocoder.
An LLM agent that runs the full ML research loop solo, from codes, trains, evaluates, and self-repairs on real recommender data.
H2.OS. The farmer's best friend, one drop at a time.
Reads every neighborhood safety report, correlates them by meaning, and wakes a human only for the ones that turn out to be a real pattern. On the demo set, 37 of 38 never surface; one alert fires.
An AI-powered museum that turns suspicious messages into interactive exhibits, exposing scam patterns and manipulation tactics before you become part of the collection.
