DBOS, Inc. Announces Technology Partnership with Databricks to Increase Agentic AI Reliability and Trust

Open source DBOS software makes agentic workflows fault-tolerant and observable at any scale

DBOS, Inc., the company pioneering open source durable execution for AI applications, announced a technology partnership with Databricks, the Data and AI company. The partnership aims to ensure AI agents provide their intended benefits to the business continuously by increasing the reliability and observability of the tasks they carry out.

AI agents are systems that autonomously carry out tasks such as processing orders, answering customer service questions, conducting research, or writing application software. Agentic AI introduces new challenges for software teams, including long-running workflows, waiting on humans-in-the-loop, and unpredictable AI model responses, all of which increase software engineering time and complexity.

The partnership brings together the DBOS open source durable workflow execution platform with the Databricks platform and Lakebase, Databricks’ serverless Postgres database built for AI agents. As AI agents execute, DBOS seamlessly stores the status of their workflows in Lakebase, in real time as workflows execute. This record of “checkpoints” ensures that AI agent workflows can resume automatically after failures, without data loss.

Also Read: AiThority Interview with Glenn Jocher, Founder & CEO, Ultralytics

“We’re excited to be partnering with Databricks; it’s a popular AI platform among our customers,” said DBOS CEO, Qian Li. “Lakebase smoothly integrates Postgres with the entire Databricks agentic stack and adding DBOS to the mix, with no extra infrastructure or coding changes required, makes Databricks-hosted agents completely resilient to failures in production.”

AI developers are already leveraging the combined platform to build large-scale agentic systems. Yutori, an AI company building autonomous agents that can take actions and complete tasks on the web, uses DBOS together with Databricks-managed Postgres to power its always-on agent workflows. “The flexibility of DBOS combined with Postgres-as-a-service from Databricks meant we could iterate quickly and still get reliability and observability right from day one,” said Abhishek Das, Co-founder and Co-CEO of Yutori.

As demand for AI-native applications accelerates, both DBOS and Databricks see a growing need for infrastructure designed specifically for agentic workloads. Industry data shows that AI agents are already driving a significant portion of modern database activity, underscoring the shift toward autonomous systems managing their own data and execution environments.

Also Read: ​​The Infrastructure War Behind the AI Boom

[To share your insights with us, please write to psen@itechseries.com ]