Want to connect with Contxts.ai?
Join organizations building the agentic web. Get introductions, share updates, and shape the future of .agent.
Is this your company?
Claim this profile to update your info, add products, and connect with the community.
Contxts.ai is a notable player in the agent ecosystem because it applies agentic design to one of the most data-dense and high-stakes environments: decentralized finance. While many agent companies focus on general productivity or chat-based interfaces, Contxts.ai uses a fleet of 40+ specialized agents to perform "infrastructure-level" tasks. These agents are responsible for autonomous vault decomposition and risk simulation, acting as a bridge between raw blockchain data and actionable financial intelligence.
The company is active in the "agentic infrastructure" and "autonomous workflows" layers of the stack. They are championing the idea that financial systems should be self-monitoring and self-executing. For those building in the agent space, Contxts.ai provides a blueprint for how agents can handle "unstructured" or multi-layered data (like complex DeFi wraps) by breaking them down into simpler components for processing. Their work in event-driven execution—where agents trigger pipelines based on on-chain signals—is a practical implementation of the "agentic loop" in a production environment.
Contxts.ai is an applied research lab based in Seoul that focuses on the engineering of financial contexts. The company emerged from NFTBank, a platform recognized for its NFT valuation models, but has since expanded its scope to address the broader complexity of decentralized finance (DeFi). The core premise of the lab is that DeFi vaults and digital asset flows have become "black boxes"—structures with multiple layers of wrapping and opaque yield sources that make risk quantification nearly impossible for human analysts or static software.
To solve this, Contxts.ai builds autonomous agents that operate at the infrastructure level. Rather than providing a passive dashboard, they have deployed more than 40 agents designed to decompose every vault into its base assets. These agents score risk at each individual layer and execute Monte Carlo simulations to predict how liquidation cascades, oracle failures, or changes in collateral regimes might impact a portfolio. The goal is to turn fragmented on-chain data into production-grade intelligence that financial workflows can use without manual intervention.
The company’s primary product delivery vehicle is Renaissance Financial. This venture represents a move toward self-running financial infrastructure. The system is designed to be event-driven: when new collateral is detected on-chain, the agents trigger a full pipeline of execution automatically. There are no manual triggers and no waiting for a daily update. This real-time, autonomous approach is a departure from the traditional "request-response" model of financial analytics. It treats financial data as a continuous flow that requires constant, agent-led monitoring and re-contextualization.
Founder and CEO Minsu Kim leads the team from South Korea, positioning the company at the intersection of the region's strong crypto-native culture and advanced engineering talent. In early 2024, the company secured approximately $12 million in funding to further develop these proprietary valuation and risk models. This capital injection is being used to expand their team of data scientists and engineers as they attempt to scale their agentic framework across more complex DeFi protocols.
Contxts.ai sits in a unique position between traditional risk management and autonomous agent development. While companies like Gauntlet or Chaos Labs provide economic simulations for protocols, Contxts.ai is focused on the "context" of the asset flows themselves—how assets move, how they are wrapped, and how they interact with different layers of the stack. Their focus on "engineering contexts" suggests they are building a foundational layer for other financial agents to build upon. By providing the underlying risk and decomposition logic, they enable a future where financial agents can make informed decisions based on the actual, unwrapped value of an asset rather than its superficial ticker symbol or protocol label.
An autonomous agent platform for DeFi vault decomposition and risk simulation.
Renaissance CLI (ren)
Anthropic-managed directory of high quality Claude Code Plugins.
Python Rest Client to interact against Schema Registry confluent server
Snowflake Connector for Python
dbt-spark contains all of the code enabling dbt to work with Apache Spark and Databricks
A framework for generating Apache Airflow DAGs from other authoring interfaces.
SQL views for Dune
Kubernetes operator for managing the lifecycle of Apache Spark applications on Kubernetes.
Singer.io Target for CSV on S3 - PipelineWise compatible
Contxts.ai is hiring
You've explored Contxts.ai.
Join organizations building the agentic web.