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:
It splits a task into stages, records the evidence behind each one, and turns every stage boundary into a checkpoint you can inspect, verify, roll back, fork, and resume from.
Give it a benchmark and walk away. An agent that does the part of ML research nobody enjoys: 40 experiments at 2am. It found a feature we'd missed, and stopped itself when it stopped learning.
Ele is a robust AI-generated image detector built to spot synthetic media even after it has been degraded by real-world manipulations like JPEG compression, blur, resizing, and noise.
An autonomous ML research agent that proposes hypotheses, edits recommender code, trains and evaluates candidates, and learns from signed results with zero manual intervention.
Adaptive GPU dispatcher achieving 3.88x geomean speedup (up to 12.01x) and streaming a 20.5 TB frontier shape at 134% MFU on a single 12 GB RTX 3060.
Turn your codebase into a living 3D clan where AI agents build, fix, and evolve your code.
Local-first AI coding harness: a Next.js control plane pairs with the clancode CLI on your machine. TrueForge runs agents locally; the clan village visualizes real RunEvents
RIFT detects AI images after compression, blur, resizing and reposting. It combines frozen CLIP with forensic signals and uses a fixed low-FPR threshold for realistic social-media evaluation.
Same outputs, 7.1x faster: hand-tuned Triton kernels and CUDA graphs for Transformers on a Tesla T4.
We gave an AI agent a dataset and a scoring rule without hints. It proposed, coded, and tested its own recommender models, which beat the official baseline entirely on its own.
A web page where people and agents who work at different times meet: state, authority and the record live on the page. Your agent comes in through WebMCP in ChatGPT. A real Codex fleet does the work.
Agent Trust Gateway enables agents to collaborate across teams without exposing private agents or files, using owner-approved, scoped, revocable, one-use access enforced and audited by middleware.
An auditable autonomous ML research loop that iterated from the officialKuaiRand-Pure FM baseline to a validation-selected DIN-lite/FM rank ensemble.
No Cost high-precision conversational e-commerce search delivering 98% hit rate in under 3 turns at sub-50ms latency.
An autonomous ML scientist that safely proposes, tests, diagnoses, and preserves recommendation-model improvements for KuaiRand-Pure.
Build a machine of ridiculous power, then deliberately push the Rift harder. SURGE future waves into the present for greater rewards. How greedy do you dare get? What dangers wait deeper in the Rift?
AI-image detectors often fail after images are compressed, resized, or reposted. InvariantGuard keeps working anyway.
An agent that talks to a shopper the way a good in-store assistant would, understand what they say, ask sharp questions, and narrows a catalog down to the right product in a handful of turns.
An offline conversational shopping agent that understands evolving preferences, handles Intent Override, and returns reliable Top-10 recommendations every turn.
Problem Statement 2: Autonomous Machine Learning Research Agent for Recommender Systems. Autonomous Research -> Code -> Evaluate loop to beat recommender baseline. Without humans in the loop.
An autonomous ML research agent for recommender systems: forms its own hypotheses, writes the code, reads the metrics, and decides what to try next.
Every generator leaves a tell. We find it and show you where
11.87× geometric-mean speedup across the 12 of 14 shapes with runnable official baselines.
An LLM agent that autonomously proposes, codes, tests, and iterates on recsys models for KuaiRand-Pure — zero manual intervention, +0.0022 delta over baseline, full run-by-run transparency.
