EVERYTHING
Datadog has significantly extended the reach of its Bits AI framework, enabling DevOps teams to automatically discover and resolve issues based on telemetry data collected by its observability platform. The new suite integrates an AI coding tool, dubbed Bits Code, across Datadog’s entire portfolio, allowing it to propose remediations and generate the necessary code to fix problems observed within the system.
Automating the Software Development Lifecycle
According to Devops, the integration of Bits is designed to address the increasing complexity and velocity of modern application development. The framework introduces several specialized agents that automate critical stages of the software lifecycle:
- Bits Code: Proposes specific remediations and generates code fixes directly from data residing in the Datadog observability platform.
- Bits Release Agent: Verifies every code change by analyzing its intended impact, generating a validation plan, running checks in staging environments, and monitoring the rollout process.
- Bits Testing Agent: Automates synthetic test generation and maintenance by exploring applications, identifying critical user journeys, and creating comprehensive test suites.
Shifting Observability to Active Governance
Beyond code management, Datadog is pushing observability into autonomous governance. Bits Remediation allows DevOps teams to invoke the AI framework to configure and run remediation scripts within strict guardrails defined by the team. Furthermore, the company is previewing several advanced capabilities that enhance operational intelligence:
- Bits Infrastructure Operations: This feature autonomously detects, investigates, and remediates common infrastructure issues based on actions previously approved by DevOps teams.
- Bits Memories: The AI framework automatically retains information from investigations, runbooks, postmortems, and Slack conversations to create and execute automated scripts.
- AI Guard Tool: Utilizes telemetry tracing to track system behavior and surface anomalous activity across the IT environment.
Datadog CEO Olivier Pomel noted that software engineers are struggling to keep pace with current code development rates, emphasizing a greater need for AI to automatically discover and remediate issues. This shift aligns with observations from industry experts like Mitch Ashley of the Futurum Group, who stated that observability is moving from merely describing production to actively governing it. The telemetry data can now be used not only to surface an incident but also to propose fixes, validate them before release, and bound the authority of autonomous agents.
The platform’s expansion also includes Bits Data Analysis, which uses the Datadog Data Context graph to allow end users to launch natural language queries to explore business metrics such as revenue, sales pipeline, and product adoption. This comprehensive suite positions Datadog at the forefront of AI-driven operational intelligence.