CockroachDB and AWS invite developers, engineers, and AI builders to create the next generation of agentic applications. Harness CockroachDB's distributed AI capabilities, fully managed MCP Server, agent-ready ccloud CLI, open-source Agent Skills Repo, LangChain integrations and Claude/Cursor plugins - all on AWS - to build AI agents with production-grade, persistent memory.
AI agents are rapidly moving from experiments into real production workflows, like writing code, running pipelines, and diagnosing incidents. But agents need memory that never goes down. An agent whose memory goes offline stops. Traditional databases were optimized for human-scale reads and writes. Agentic systems spawn autonomously, write constantly, and require memory that persists across regions, failures, and scale. CockroachDB was built for exactly this. It is the system of record for agentic memory: globally distributed, always-on, PostgreSQL-compatible, and now natively integrated into the agent toolchain through MCP, cloud, and an open-source skills ecosystem.
The Challenge Build an agentic application that uses CockroachDB as its persistent memory layer, deployed on AWS. Your agent should store, retrieve, and act on memory whether that's conversation history, user context, task state, embeddings, or structured transactional data. The best submissions will demonstrate that memory is not an afterthought, it is the thing that makes an agent useful in production.
All submissions must use at least two CockroachDB tools (MCP Server, Distributed Vector Indexing, ccloud CLI, or Agent Skills Repo) and at least one AWS service (Amazon Bedrock, AWS Lambda, Amazon ECS / EKS, Amazon S3, Amazon SageMaker, or Amazon Bedrock Agents).
What to Submit - Provide a URL to your public open source code repository with clear README documentation and an open source license file (MIT or Apache 2.0). - Provide a URL to your functional demo app. - Include a video (less than 3 minutes) demonstrating your submission and the CockroachDB memory layer at work. - Identify which CockroachDB and AWS tools you used and how. - Optional: Include an architectural diagram or feedback on the tools.
Prizes - 1st Place: $5,000 in USD, Blog feature, Cockroach Labs Swag - 2nd Place: $2,500 in USD, Cockroach Labs Swag - 3rd Place: $1,250 in USD, Cockroach Labs Swag
Judging Criteria - Agentic Memory Design: Does CockroachDB play a meaningful, production-grade role as the agent's memory layer? - Technical Implementation: Is the integration with CockroachDB tools quality software engineering? - Real-World Impact: How big of an impact could the project have on real users or workflows? - Production Readiness: Is the design secure, observable, and scalable? - Creativity & Originality: Is this a genuinely new idea or a novel application of the technology?