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:
We built a 21.7M-parameter ViT detector for robust AI-image detection, scoring 0.9362 across unseen generators, with deployment via Chrome extension and future TikTok integration.
Most AI-image detectors ace clean benchmarks and collapse on real uploads. Ours holds 0.99 AUROC across every compression, blur, crop and noise level tested, and 0.95 on a generator it has never seen.
AI-image detectors often fail after images are compressed, resized, or reposted. InvariantGuard keeps working anyway.
RecommenderForge autonomously tests ranking ideas, learns from validation, recovers from failures, and ships checkpoint-backed predictions—making recommender research fast, safe, and easily traceable.
A deterministic 2-level Bayesian shopping agent achieving 0.9744 TechnicalScore across 50k products with zero LLM calls, sub-15ms latency, and expected utility depth.
A safety-gated five-agent system that autonomously proposes, runs, evaluates, and remembers recommender experiments—without exposing the hidden test set.
An AI-image detector that compares an input against real prototypes. Our model performs excellent against multiple transformations, but needs generator specific data to scale.
An LLM-driven agent that autonomously writes, runs, and debugs its own code to improve a recommender ranking model — reading its own errors, banning failed approaches, and iterating without human help
An autonomous AI housing agent that understands your needs, finds and ranks the best properties, and continuously monitors the market for new matches.
Harness buddy automates all the ML research for you. It writes training code, reads its own crashes, and critiques itself when stuck. Beat the FM baseline on hidden test. Zero human interventions.
An e-commerce copilot that turns conversation into ranked products, with a self-iterating prompt lab that evaluates and refines intent understanding without fine-tuning.
R&D that doesn't sleep: an autonomous agent that improves your recommender overnight while you zzz
“AI-powered farming, smarter decisions, better harvests.”
Your autonomous AI production manager for turning screenplays into smarter, production-ready plans.
An enterprise grade serverless AI Mentor utilizing multi agent alignment frameworks to deliver ecologically conscious safe science education globally
CrowdSense uses AI, CCTV and IoT/RF sensing to predict mall crowd and queue buildup before it happens, helping operators act early and users make smarter decisions.
An enterprise-grade serverless AI Mentor utilizing optimized multi-agent alignment frameworks to deliver hallucination-free, ethically sandboxed EdTech tutoring globally.
AI Clinical Copilot connecting MRI scans, blood reports, consultations, prescriptions, and surgery with intelligent analysis, voice assistance, and real-time clinical decision support.
Drop-in PyTorch Transformer layer for Track 3: 1.92× geo-mean speedup, strict state_dict, 932× under tolerance. Shape 14 completes in 5 GB; the reference cannot run (18 TB score tensor per layer).
InboxPilot is an autonomous AI email triage assistant that classifies emails, prioritizes urgency, drafts responses, syncs actionable items to Notion, and learns user preferences from feedback.
AI image detector
Building production-ready e-learning content usually requires manual coordination across content generation, syllabus structuring, and task prioritization.
A small autonomous loop for KuaiRand-Pure within-user ranking. It reproduces the official FM, tries one change at a time on validation only, then retrains and writes the test scores.
Can you tell if an image was made by AI?Real or Fake (ROF) is a model that looks for AI-generated images, even when they've been blurred, resized, cropped, or heavily compressed.
