Developers lose hours each week to a familiar troubleshooting loop: chase down telemetry across dashboards, decipher vague errors, and juggle alerts to find the signal worth fixing. Production issues, performance regressions, and security vulnerabilities all demand attention, but they often come with little context for taking action.

To win back developers’ time, the Bits AI Dev Agent is Datadog’s AI-powered coding agent that autonomously monitors your observability data, identifies high-impact issues, and opens production-ready pull requests (PRs) complete with tests, context, and human-readable explanations. With the Bits AI Dev Agent, developers can shift their focus from investigating problems to reviewing fixes and building code. This is one of three AI agents designed to support roles across development, SRE, and security.

In this post, we’ll cover how the Bits AI Dev Agent:

• Understands and acts on all your observability data directly in Datadog
• Works autonomously to generate fixes and deliver asynchronous PRs for high-impact issues
• Investigates and generates code fixes across the Datadog platform

Understands and acts on all your observability data

Unlike stand-alone coding copilots, the Bits AI Dev Agent is fully embedded in Datadog. It ingests logs, traces, metrics, Real User Monitoring (RUM) events, security signals, and runtime variables in real time, allowing it to build a rich production context to diagnose the problem before suggesting any changes to your codebase. After evaluating multiple possible solutions, it writes a fix that matches your team’s style guide, applies formatting, and prepares the diff for review.

From there, a user can decide whether to create a PR based on a potential fix. The Bits AI Dev Agent watches CI logs—either via Datadog CI Visibility or GitHub Actions—and iterates automatically until tests pass. The result is a production-ready PR that’s aligned with your team’s code quality and standards.

By combining observability with code awareness, the Bits AI Dev Agent automatically triages noisy telemetry data to surface the most impactful issues and generates context-rich, test-backed patches grounded in real system behavior. It tracks every action within the same platform where teams already monitor and alert, so your developers don’t need to switch tools or hunt for root causes; they can immediately get accurate diagnoses and tested solutions.

Works autonomously to generate fixes and delivers asynchronous PRs for high-impact issues

Fixing code issues can be time-consuming, especially when it’s unclear which ones need action and which can be safely ignored. To help teams triage and resolve problems more efficiently, the Bits AI Dev Agent offers autonomous code fixes within Error Tracking.

This feature works behind the scenes to analyze new issues, assess their impact, and—when possible—generate a code fix or explanation. In many cases, the Bits AI Dev Agent can surface a fully scoped explanation and suggested fix at the moment an engineer opens the error in the UI, eliminating the need for manual investigation. For minor or ignorable issues, the Bits AI Dev Agent explains why no fix is needed, helping reduce noise and avoid unnecessary PRs.

For high-impact issues—such as crashes, 500 errors, or anything affecting a large number of users—the Bits AI Dev Agent can go a step further: It will generate a PR with the fix pre-applied. This enables engineers to immediately review and merge a resolution without having to start from scratch. Whether or not a PR is created, the Bits AI Dev Agent’s background analysis ensures that important problems are ready to act on by the time they’re surfaced in the UI.

Automates code fixes across the Datadog platform

The Bits AI Dev Agent supports a wide range of issue types across Datadog products. It’s generally available in Error Tracking, and in preview for APM, Proactive App Recommendations, Test Optimization, Code Security, and Continuous Profiler. Database Monitoring support is coming soon, with the ability to optimize slow queries and remove inefficient loops.

Regardless of the originating product, each PR generated by the Bits AI Dev Agent includes full context and test coverage.

Get started with the Bits AI Dev Agent

To start using the Bits AI Dev Agent:

1. Install the Datadog GitHub App with appropriate permissions.
2. Ensure your services are tagged with service and version.
3. Enable the Bits AI Dev Agent from the Error Tracking page in Datadog.

Once enabled, the Bits AI Dev Agent will automatically begin triaging issues and suggesting fixes across supported products, eliminating the friction between detection and resolution. It identifies issues in real time, writes production-ready fixes, and opens PRs on your behalf—all while giving your team members the visibility and control they need to stay in the loop.

We’re continuing to expand the Bits AI Dev Agent, with upcoming support for GitLab and Azure Repos, a secure sandbox for build-and-test iterations, and expanded coverage across more Datadog products. We will also soon include an Autonomous Mode setting for Error Tracking that will give users more control over how the Bits AI Dev Agent autonomously solves issues.

Start using the Bits AI Dev Agent today to spend less time triaging and more time building, and sign up for the Preview to get access to platform-wide support. If you’re new to Datadog, sign up for a 14-day free trial.