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
Making chest X-ray models interpretable through SPN-guided counterfactual explanations.
Gemedi uses dual-discriminator feedback to generate synthetic discharge notes that balance linguistic realism and medical reasoning.
Deeplearning model that automatically separates piano MIDI recordings into left-hand and right-hand tracks
MedShift: Robust ICU Mortality Prediction Across Hospital Systems
We built a classification model that helps to recognize adversarial patterns in images and differentiate clean and adversarial examples.
Our project introduces the Depth-Aware Relevance Transformer (DART), a model designed to predict how important different regions of an image are by combining RGB, depth, and spatial information.
n/a
I will develop a graph neural network that integrates multimodal genomic data to learn external factors that guide the binding of the dosage compensation complex on the X-chromosome vs. autosomes.
Transforming Drug-Target Interaction Prediction: Augmenting Top-DTI with Graph Transformers
Final project
The goal of this project is to empower typing-based biometrics authentication, where a user may be verified based on how they type a phrase.
Prior research demonstrates that even with balanced datasets, deep neural networks amplify gender biases due to correlated features. We will extend adversarial debiasing to image captioning tasks.
I aim to provide musical analysis of how models understand genre-distinguishing features, with a focus on hybrid genres like Afrobeats that blend with contemporary hip-hop and electronic elements.
We investigate how LLMs can be manipulated to produce harmful responses using visual adversarial examples.
Making a model to maximize single channel ECG to continuously predict blood pressure.
？
We explore whether vision transformers can perform abstract reasoning tasks using by solving puzzles. CLIP vs. DiNOv3
Using Fin-bert and LSTM to predict the stock market
A multimodal AI system that learns user intent by combining gesture recognition with facial-expression-driven emotional feedback.
Making unique molecules out of CHOND with a Sequential VAE
Our project is a re-implementation of a Minesweeper Agent from “Training a Minesweeper Agent Using a Convolutional Neural Network”.
Audio feature extraction though AST & LeJEPA
...
Tiny models that pack a big punch.
1 – 24 of 33
