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
Peaview is an AI research project that ran 102 experiments in 72 hours, that took it's time to learn from it's mistake and explore different angles to tackle the problem, beating the baseline by 16%.
CartChatter asks the right questions to find the right product — even when you change your mind.
SLB-ScopedRun turns "don't look there" into "it isn't there." Each Agent Run mounts only the one delegated resource, read-only, enforced server-side — not by asking the model nicely.
TikTok TechJam「Autonomous ML Research Agent for Recommender Systems」
Finds that 1 in 50,000 by remembering & showing its discoveries, every step of the way
Afterimage detects AI-generated images after the platform has compressed, blurred and resized them — cutting accuracy loss under real-world post-processing from 15.7 points to 2.5.
An offline-first conversational shopping agent that turns changing preferences into fast, explainable, catalog-valid recommendations—without network calls or model tokens.
An autonomous ML scientist that safely proposes, tests, diagnoses, and preserves recommendation-model improvements for KuaiRand-Pure.
An auditable autonomous ML research loop that iterated from the officialKuaiRand-Pure FM baseline to a validation-selected DIN-lite/FM rank ensemble.
Conversational shopping search that remembers, adapts, and recommends fast.
Sentinel governs AI agent actions in real time—allowing safe work, holding risky commands for human approval, and blocking threats to protect your data and systems.
A verbatim-evidence shopping agent that finds the hidden target among 50,000 products in ~2.5 turns: 0.962 vs 0.107 baseline, HR@10 = 1.000, deterministic and fully offline.
A robust dual-stream AI image detector fusing DINOv2 and SRM steganalysis to deliver calibrated forensic scores invariant to real-world redistribution degradations.
An autonomous ML research agent for recommender systems that tries to disprove its own findings. It found a +0.0314 result, proved it was worth +0.0023, and reported the smaller number.
"Volc Agent Launchpad: an AI agent platform where every file access is authorized server-side with scoped, revocable keycards
0.6075 vs the official 0.6016 baseline on KuaiRand-Pure — ~7× seed noise, CPU-only, inside one 300s budget — produced autonomously: 5 iterations to convergence, 0 manual interventions.
...but what if we could put a leash on our agents?
Cereal Crunch is a binary AI-generated-image classifier (<2B parameters) that provides detection on both clean images as well as on images under six post-processing image transforms.
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 safety-gated five-agent system that autonomously proposes, runs, evaluates, and remembers recommender experiments—without exposing the hidden test set.
A shopping agent that asks before it guesses. Searches 50,000 products with no model, no API, no network - Python standard library only. Scores 0.963 on the organizer's evaluator.
Your shopping soul mate: two questions, and it pulls the exact product out of your head — from 50,000 items, zero LLM, MRR 1.0.
Shoply, your personal AI shopping assistant. Tell it what you are looking for and it will help you find exactly what you need!
Like Dory finding Nemo, Agent Dory searches a sea of products through conversation, using evolving clues and follow-up questions to find the hidden target.
289 – 312 of 614
