AI Coding Agents Drive New Observability Revenue for Datadog, Dynatrace

AI agents may expand the monitoring market

The rise of AI coding agents is creating an unexpected opportunity for observability platforms. Rather than reducing the need for monitoring by writing better code, autonomous agents are generating more software changes, deployment events, and production complexity that companies need to track.

Datadog and Dynatrace are both moving to capture this emerging workload, though with different approaches and momentum profiles.

Datadog frames the inference economy

At the Goldman Sachs Communacopia + Technology Conference on September 10, Datadog disclosed that thousands of customers are already using its AI-agent observability capabilities and that monetization has begun. Management described an "inference economy" in which proliferating AI applications and agents create continuous monitoring demand as software makes decisions and modifies itself—shifting the focus from tracking human-authored releases to tracing machine-driven activity.

The company reported second-quarter revenue of $1.12 billion, up 36% year-over-year, with customers generating at least $100,000 in annual recurring revenue climbing 23% to approximately 4,720. Datadog's usage-based pricing model positions it to capture expanding telemetry volume as agents generate more events. The trade-off is exposure to optimization pressure if more capable agents reduce incident frequency.

Dynatrace integrates directly into agent workflows

One day after Datadog's conference remarks, Dynatrace launched a verified plugin for Cursor Marketplace. The integration delivers live production context and 30 Dynatrace skills directly to coding agents through a single installation, embedding observability data into the agent's decision-making process.

Dynatrace prices the Cursor integration on data consumption rather than per developer seat, insulating it from potential declines in human headcount. The company's annualized recurring revenue grew 17% to $2.136 billion in its latest quarter, while annualized log consumption reached $200 million after nearly doubling over two quarters. However, Dynatrace's slower overall growth and its planned $915 million acquisition of Arize introduce execution risk.

Why it matters

The observability market faces a fundamental question: will AI agents prevent enough problems to shrink monitoring needs, or will they generate enough autonomous software activity to make observability a larger machine-to-machine business? Early evidence suggests the latter. If agents continuously modify code, trigger deployments, and make runtime decisions, the volume of events to trace could dwarf today's human-driven release cycles—even as developer teams shrink. That dynamic would favor usage-based revenue models and vendors that can instrument agent-to-agent interactions, not just traditional application performance.

Institutional positioning

According to Insider Monkey's hedge fund database, 92 hedge funds held Datadog shares in the second quarter of 2026, up from 80 in the first quarter; Arrowstreet Capital increased its position 40% to 2,195,554 shares. Dynatrace holders rose to 48 from 46, with AQR Capital Management boosting its stake 346% to 8,317,467 shares. As of August 31, short interest in Datadog stood at 10,220,303 shares, representing 3.04% of float and 2.94 days of average volume.

These details were first reported by AI Watch.