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
Real-world synthetic image detection built on frozen visual representations
Shopping Copilot - Type what you want
AI image detection that stays accurate even when images get compressed, blurred, or resized.
Real Or Reel Is Now a Question of The Past
AI can fake reality. We built a detector that doesn’t just work in the lab—it stays accurate when AI images are edited, compressed, and shared online.
Conversational shopping that understands when a customer has a clear preference, when they’re browsing, and when it’s smarter to ask a helpful question instead of guessing.
For TechJam Track 4
The rise of AI is making it increasingly difficult to differentiate AI generated versus authentic images. What if we could use AI to beat it at its own game? Introducing our very own AI that can!
WOW-WILDLOTUS Shopping Copilot understands evolving needs, searches 50,000 products, asks smart questions, and dynamically adapts recommendations to help shoppers find the right product faster.
A zero-trust agent middleware enforcing tenant isolation, prompt-injection defense, secret redaction, and human approval gates for safe execution.
Filter on nothing. Score on everything.
The shopping copilot that evolves itself.
We built AURA, an Adaptive User-Reasoning Agent for Conversational E-Commerce. AURA achieves a 94.0% Hit Rate on 50,000 Amazon products.
A conversational shopping agent that asks smarter questions and finds the right product faster.
An autonomous, latency-optimized conversational shopping agent leveraging in-memory SQLite FTS5 retrieval to discover a customer's hidden target product within 10 turns.
Real products, real conversations
Shape-aware Triton kernels that fuse Transformer attention into tiled, memory-lean GPU launches. 1.977× faster inference on RTX 5070 Ti, zero failed elements across 938M+ outputs.
CompassCart is an offline AI product-search agent using a versioned ledger, hybrid retrieval, and adaptive reranking. It asks only high-gain questions and guarantees 10 sanitized, valid catalog IDs.
An AI research engineer that improves recommender models while you sleep.
because seeing shouldn't mean believing
RobustLens estimates whether an image is AI-generated after layers of post-processing using calibrated multi-view scoring, transformation-chain testing, and safe abstention when evidence degrades.
Zero-Cost, Sub-15 ms Conversational Search for Modern E-Commerce
An offline conversational search agent that tracks changing shopper intent and ranks products from a 50K-item catalog using FTS5 and weighted BM25. It achieved a 96% Hit Rate@10.
TICK TOCK, time is compute ⏱️ Baseline? Just the starting line 🏁
529 – 552 of 614
