Datris has announced a significant expansion of its agent-native data platform, moving beyond traditional chat interfaces to make AI agents "first-class operators" of data infrastructure. Built around the Model Context Protocol (MCP), the platform allows AI agents to autonomously connect to data sources, build ingestion pipelines from scratch, and manage their own security credentials without human intervention. By providing agents with the tools typically reserved for data engineers—such as ingestion, validation, and observability—Datris aims to eliminate the "glue code" often required to connect AI to enterprise systems.
Datris enables AI agents to autonomously build, manage, and observe data pipelines.
Built entirely on the Model Context Protocol (MCP) for seamless agent integration.
Agents can now connect to S3, databases, and enterprise APIs (Workday, Salesforce, ServiceNow).
The platform introduces "taps"—on-demand or recurring data feeds defined in plain English.
Features a dedicated security layer where agents manage their own scoped credentials.
Open Source: Released under AGPL-3.0 and available for self-hosting via Docker.
While much of the industry has focused on adding "copilots" to existing software, Datris has rebuilt the data stack specifically for autonomous agents. This "agent-native" approach allows an agent to describe a task—such as normalizing timezones or dropping malformed rows—and have the platform automatically generate the schema and transformation logic.
"The data industry spent twenty years building tools for human engineers, and the last two trying to retrofit them for AI," said Todd Fearn, founder of Datris. "An agent should be able to land on the platform, set up its own credentials, build a pipeline, and run it... without a human in the loop."
To ensure safety in an autonomous environment, Datris enforces a strict line between human and agent credentials. Agents can request, rotate, and delete API keys scoped to their own tasks, while human-owned credentials remain untouchable. Additionally, a live operations view provides real-time observability, translating technical errors into language that the agent can understand and act upon to self-correct.
The platform is fully open-source and self-hostable, allowing teams to inspect every line of code. For organizations that prefer not to manage their own infrastructure, a hosted version is available at datris.ai.
About Datris
Datris is the first AI agent-native data platform. It gives AI agents the same tools human data engineers use — for ingestion, validation, transformation, search, and observability — through a single MCP interface.
TechIntelPro empowers B2B technology decision-makers with the latest industry trends, deep insights, leadership perspectives and actionable intelligence to drive both organizational and professional milestones—and thrive on the tech pulse.