Datadog unveils new Bits AI agents to streamline cloud workflows

Datadog has announced the introduction of three artificial intelligence agents designed to support development, security and operations teams with interactive investigations and asynchronous code fixes.

The newly released Bits AI SRE, Bits AI Dev Agent and Bits AI Security Analyst agents are complemented by Proactive App Recommendations and APM Investigator capabilities, furthering the development of Bits AI, Datadog's generative AI assistant intended for real-time resolution of application issues.

Domain-specific AI agents

Datadog has stated that these new domain-specific AI agents are trained to provide expertise in incident response, product development and security for organisations operating cloud applications. Each agent reportedly operates on a shared set of core capabilities, enabling tasks such as querying data, analysing anomalies, and scaling infrastructure to be consistently deployed and reused across different agents.

This architecture is aimed at allowing the rapid creation and deployment of new agents, whilst maintaining a uniform user experience. According to the company, this framework is supported by substantial and high-quality observability data, ensuring the AI capabilities work with contextual understanding and accuracy to deliver actionable insights intended to eliminate risk.

"Datadog is uniquely positioned to deliver value with AI as a platform that has a wealth of clean, rich data—we process trillions of data points and are embedded in our customers' critical engineering, developer and security workflows," said Yanbing Li, Chief Product Officer at Datadog. "With these advancements in AI reasoning and multi-modality, we've gone beyond helping organisations understand their availability, security, performance and reliability. We now enable human-in-the-middle workflows by guiding customers on what to look for and where to start looking, and augment their ability to take action."

Agent details

The Bits AI SRE agent, now in Limited Availability, functions as a 24x7 on-call responder. The agent undertakes the initial triage of all alerts using telemetry and service context to produce preliminary investigation findings prior to an individual responder logging on. It then assigns relevant roles, provides updates through real-time incident summaries and status reports, suggests subsequent steps, and drafts initial versions of incident post-mortems to save time for response teams.

The Bits AI Dev Agent, available in preview, is designed to detect software issues, propose code fixes, and directly open pull requests tailored to organisations' technology stack. This feature aims to streamline the review and merge process within source code management systems, providing engineering teams with AI support for addressing issues and enhancing productivity.

Bits AI Security Analyst, also in preview, performs autonomous triage of Cloud SIEM (security information and event management) signals. The agent is intended to handle in-depth analysis of potential threats and deliver well-reasoned resolution recommendations without requiring human intervention, potentially reducing incident response times and transforming the security workflow involved in processing security signals.

Customer commentary

Thomson Reuters is among the organisations adopting Datadog's AI offerings. Darren Trzynka, Senior Cloud Architect at Thomson Reuters, shared the company's perspective on the benefits of Bits AI for operational teams.

"At Thomson Reuters, we're focused on maximising operational efficiency and accelerating innovation at scale through generative AI solutions," said Darren Trzynka, Senior Cloud Architect at Thomson Reuters. "Bits AI allows operations and downstream platform teams to receive the full context of the investigation—from the initial monitor trigger to conclusion—driving down resolution time significantly freeing them up to do more."

Additional capabilities

Datadog has also introduced Proactive App Recommendations and APM Investigator features as part of its preview capabilities. Proactive App Recommendations uses telemetry data already collected by Datadog to suggest impactful fixes and recommend next steps. This tool provides insights to developers for optimising slow queries, addressing inefficient code, and catching recurring exceptions, with the goal of improving performance, reducing errors, and cutting resource usage before end-users are affected.

The APM Investigator tool automates the previously manual task of troubleshooting latency spikes. It works by identifying bottlenecks, gauging impact, highlighting trends across slow traces, suggesting possible causes, and proposing resolutions.

Other launches referenced include capabilities in AI observability, AI security, log management and the rollout of an internal developer portal. The newly announced AI solutions are designed to complement Datadog's ongoing commitment to support engineering, security, and operational workflows within cloud environments.