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
CLIP-B last-4 (~86M) scores 0.990 on 0.50×clean AUC + 0.50×14 social transforms. A mean-logit fuse with a mixed-generator head reaches 0.993. Official val and EvalGEN never enter training.
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.
A Kill Switch for AI Agents that blocks work a task never authorised — not because the action looks dangerous, but because this run wasn't allowed to do it.
An ML research agent that owns the whole loop — it forms the hypothesis, writes the code, trains, and grades itself. 23 experiments, zero human code, and it caught its own data leak before we did.
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.
We built an autonomous ML engineer that forms hypotheses, writes and tests models, learns from failures, and iteratively improves recommender system
A shopping copilot that treats every product it shows as a commitment rather than a list, and every question it asks as a turn it has to earn.
"Dont put keys in agent context" WATCH ME. Now with sandboxxing and storage for codified processes.
An LLM agent that diagnoses, changes, retrains and measures a recommendation model in a loop: 0.6016 → 0.6056 on KuaiRand-Pure, no human intervention, and a harness built to check its work.
A shopping agent that knows the difference between browsing and buying - and asks smarter questions because of it.
An AI Agent with direct access to a company's documents has more authority than it needs. VaultGate enforces need-to-know access before any protected content ever reaches the model.
A deterministic, fully-offline conversational shopping agent that finds the exact product in under 3 turns — 7.7× the official baseline, zero LLM tokens.
Running attention inference on 512 KB of sram on an ESP32
Team Senior Prompt Engineers working on creating an Autonomous Machine Learning Research Agent for Recommendation system, using the KuaiRand-Pure dataset.
An autonomous multi-agent AI research team that invents, writes, tests, and self-heals recommendation system models directly in PyTorch code.
Stop searching. Start shopping through conversation. Shopping Clerk asks, remembers, and adapts to find the right products as your intent evolves.
Every Agent Run is guilty until proven safe.
A universal autonomous RecSys research agent with adversarial peer-review & 4-pillar taxonomy that iteratively invents, optimizes, and converges PyTorch models.
An LLM agent that does ML research unattended: 20 iterations, zero human intervention, six compounding discoveries, and +0.0043 over TikTok KuaiRand-Pure baseline on the hidden test.
FlowState: An autonomous ML agentic workflow that diagnoses data, iterates pipeline code, and optimizes recommender systems end-to-end by competing against itself, to make machine learning easier.
Shoppers don't describe products, they quote them. Copaon matches the quote back to the catalog: 50,000 products narrowed to one in two turns.
Agent middleware that writes the security policy by watching the work: recurring tasks become a specialist Agent with a distilled brief and a scope derived from what its runs really did.
A shopping copilot that asks the right question instead of showing the wrong 10,000 results. It routes buying vs. browsing intent, then converges on the exact product in a handful of turns.
Autonomous, leakage-safe ML research for competition-ready recommendations.
169 – 192 of 614
