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
An autonomous ML-research agent: +0.0040 over baseline, 45 runs, $2.55, zero interventions
Shopping Copilot is a headless conversational product-search Agent that turns a static query into a shopping process users can correct, built for the TikTok TechJam 2026 Shopping Copilot challenge.
search gives you ten blue shoes. needle asks the question that makes nine of them irrelevant.
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.
SMIFF-ing out AI-generated images across real-world edits and compression.
ByteRush unleashes autonomous AI agents to build smarter recommenders—turning raw behavior data into tested ideas, faster experiments, and personalized experiences with minimal human effort.
AegisGraph gives AI agents run-scoped permissions, owner isolation, policy checks, auditable traces, and an exact kill switch—so untrusted content never gains unchecked authority.
Ask better. Rank earlier. Adapt when intent changes.
Same math. Fewer kernels. Less memory
Adaptive, durable middleware that makes multi-Agent work visible, recoverable, and safe.
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.
An intent-aware conversational shopping agent that remembers evolving customer needs, asks high-value questions, and combines exact, lexical, and semantic retrieval to return a better Top 10.
A robust ConvNeXt-Tiny deepfake detector achieving 98.9% accuracy and 0.9994 AUC, engineered with physics-informed augmentation to survive JPEG compression and image degradation.
AI agents can retry conversations, but retrying real-world actions, like bookings or payments, can accidentally create duplicates. Our agent Action Ledger makes those actions crash-safe.
ContextCart turns evolving shopping conversations into fast, explainable recommendations using offline-safe retrieval, intent routing, and preference-aware ranking.
Physics-based forensic features + Clip, combined by an interpretable random forest: A robust, calibrated AI-image detector that can say WHY an image looks fake, and stay fast when it must.
RobustFusion detects AI-generated images reliably—even after real-world edits and transformations.
No Cost high-precision conversational e-commerce search delivering 98% hit rate in under 3 turns at sub-50ms latency.
From oops to SOTA—an autonomous ML scientist that turns failed experiments, weird metrics, and fresh research into better models, one loop at a time.
Make an AI agent run understandable: see what the model thought, what commands it ran, and exactly where it failed with secrets scrubbed before they ever hit a log.
ProofLens detects AI-generated images while remaining robust to real-world transformations such as compression, blur, cropping, color changes, and resizing.
An autonomous ML research agent. It verifies its own results instead of trusting them, and beats the baseline where a greedy agent scores below it.
Fully offline conversational shopping agent that finds a customer’s hidden target product inside a frozen 50,000-product Amazon catalog within at most 10 dialogue turns.
Agent Launchpad provides an identity-first middleware layer for secure multi-agent systems, enforcing authentication, authorization, ownership, collaboration, and auditability across Agent actions.
145 – 168 of 614
