DigitalOcean Launches New AI Agent Management Tools

The services enable developers to identify performance gaps and test model refinements through simulated environments.

Updated on Oct. 26, 2026 in Artificial Intelligence

Isometric editorial illustration showing a modular data-architecture model composed of server blocks and geometric pathways, representing software performance simulation. — DigitalOcean launched new AI agent management tools, enabling developers to simulate model refinements and track performance through iterative testing services. AI Illustration. Upload story photo >

Live Poll

Would you adopt AI agent automation tools to identify and fix software bugs automatically?

DigitalOcean has introduced new tools for AI agent lifecycle management, including Learning Services and Agent Data Services. These tools, currently in public and private preview, enable developers to track agent performance, identify failures, and refine models through simulation.

Why it matters

As AI agents move beyond simple automation to iterative tasks, the ability to close the loop between agent failures and model refinement becomes critical. This update provides infrastructure for developers to debug and iterate on agent behavior using their own interaction data.

The new suite enables developers to identify and log performance gaps, such as the 400 user sessions where failures were recorded. These metrics are then processed and indexed through Agent Data Services to facilitate targeted model improvement.

The players

DigitalOcean

A cloud computing infrastructure provider focused on simplifying deployment for developers, startups, and small to medium-sized businesses.

The details

DigitalOcean launched Learning Services, which uses a trio of tools to manage the agent lifecycle: Insights for recording activity, Signals for detecting performance gaps, and Simulations for validating updates. Signals function by archiving session data into Agent Data Services—a repository for logs, documents, and media—which then informs the Inference Router. The Inference Router acts as an optimization layer that selects the most efficient model for a specific task based on these historical performance insights.

October 26, 2026: DigitalOcean announced the new AI agent services.

The Tech Race

The new update extends the capabilities of the Inference Router by integrating it with persistent data services, marking a shift toward stateful agent management. This development places DigitalOcean’s tooling in direct competition with emerging agent-centric MLOps platforms.

Developers using the DigitalOcean cloud platform can now opt into these preview services to test agent workflows. The tools integrate directly into existing deployments, allowing for immediate observation of model performance and session logs.

The takeaway

DigitalOcean is positioning its platform to handle the full lifecycle of AI agents rather than just raw compute. Users should monitor the transition from preview to general availability to determine how these integrated simulation tools fit into current MLOps cost structures.