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:
Revolutionizing drone control with wireless local detection and GPS tracking at a very low cost
Zephyr helps vehicle designers see airflow, explore lower-drag shapes, and understand every change with NVIDIA DoMINO predictions and Nemotron reasoning served by Nebius.
Stop guessing why videos go viral, VIOS reverse engineers 10,000+ videos into instant visual memory search, real retention blueprints, and hidden content gaps competitors missed.
Fully offline AI backpack for the blind with haptic obstacle warning and voice assistant
AI that spots an attack on CCTV, dispatches help over iMessage and warns everyone nearby, even offline.
Street Accessibility and Guidance Audit
An Edge-AI platform for simulating and monitoring at-risk nursing home and dementia patients in real-time.
Hearify turns a phone video of a room into a 3D sound map, so blind people can rehearse an unfamiliar place at home, by ear, before they arrive.
AquaVision uses AI to identify fish species and estimate biomass, helping aquaculture farmers improve stock assessment, feed planning, and farm management.
Turning Apertus 8B into an unsupervised cross-lingual semantic-difference detector by reading its middle layer hidden states and mean-centering them with no prompting, no training.
Due to habitat loss, over 50% of bat species urgently need conservation action. Aiding citizen science efforts, telemetry data and computer vision is used to find improvements to bat roosts.
A GPU genetic algorithm evolves which order-book signals XGBoost uses to predict whether bursts of aggressive trading in S&P 500 futures will snap back, net of trading costs.
Every mother’s story deserves to be remembered, connected, and understood—wherever care begins.
Sanad turns handwritten maternal and newborn health forms into digital records with offline AI and human review, helping midwives spend less time on paperwork and more time on care.
InferQueue is an adaptive LLM inference scheduler for constrained local GPUs. Predicts how expensive each request will be, then uses estimate to reduce blocking while preventing requests from starving
We taught AI to read the Fed without peeking at the future: every speech, every press conference, even Powell's eyebrows, turned into trades on bonds and the dollar.
Do you know what you need in a heist? With this all you need is One Briefcase!
We trained a neural net on our own racing laps, handed it a real steering wheel, and watched it rip. Embodied AI that learns from pixels and turns a physical wheel to race in a driving game - FH4
A bio-inspired dual-process cognitive controller that gives open-weight AI human-like deliberate reasoning, zero catastrophic forgetting, and 50% compute savings on consumer hardware.
An autonomous AI research engineer that turns a research question into experiments, trains/finetunes models on GPUs, evaluates them, discovers failures, and produces reproducible research reports.
I trained ODIN Pocket from scratch on an RTX 4070. It’s a 49.3M-parameter language model you can run on a CPU to generate text and inspect the probability of each token.
A 49.5M-parameter language model trained from scratch for $14 of GPU time that beats GPT-2-124M on ARC-Easy. Tells you plainly which of its other results are just noise.
HelixDepth tests whether reusing six Transformer blocks across twelve stages can improve next-token prediction in a 41-million-parameter language model.
Convert videos into photorealistic 3D Gaussian & Triangle Splats with an embodied Spatial AI Guide powered by NVIDIA Nemotron on Nebius Token Factory and Tavily web grounding in WebXR.
