Datadog | The Monitor blog

From traces to experiments: A loop for improving AI agents

To effectively optimize AI agentic systems, teams should use aggregate trace data to identify specific performance bottlenecks and formulate testable hypotheses. The article propos…

Making Rust observability reliable at scale with OpenTelemetry

To resolve fragmented and inconsistent observability in their Rust services, Datadog developed `dd-trace-rs`, an opinionated tracer built on OpenTelemetry that ensures consistent c…

Troubleshoot and secure your code faster with Datadog’s Bitbucket Cloud Source Code integration

The Datadog Bitbucket Cloud Source Code integration connects Bitbucket repositories to Datadog products to streamline issue investigation and security monitoring. It accelerates tr…

Respond to security threats faster with Tines and Observability Pipelines

The integration of Tines and Datadog Observability Pipelines enables security teams to automate the enrichment and filtering of log data in real time. By using Tines to drive autom…

How Bits Database Optimization proves a query rewrite is faster

Datadog’s Bits Database Optimization feature provides a controlled, disposable environment to benchmark query rewrites using real production schemas and representative synthetic da…

Reduce sensitive data exposure with build-time allowlists

Datadog’s build-time allowlisting feature enables engineering teams to balance data privacy with observability by protecting sensitive, runtime-generated information while preservi…

What we learned about AI agent security by monitoring our agents

To effectively secure AI agents, organizations must move beyond monitoring final API calls and instead implement a comprehensive "AI Bill of Materials" (AI-BOM) that inventories al…

Debug live production code without redeploying with Datadog Live Debugger

Datadog’s Live Debugger allows developers to investigate production bugs in real-time without the need for code changes or redeployments by using non-breaking logpoints to capture …

Beyond the $1 AI era: How federal agencies can build the evidence for FY27 renewals

As low-cost OneGov AI promotions approach expiration in 2026, federal agencies must evaluate their platform usage and costs to prepare for FY27 renewals. Datadog Cloud Cost Managem…

How Datadog saves over $1 million each month by optimizing AI usage

Datadog achieved over $1 million in monthly AI cost savings by "rightsizing" model configurations, such as migrating from Claude Opus to Sonnet and adjusting tool effort levels. By…

Golden Paths for AI agents: What changes when platform users aren’t human?

To accommodate the rise of AI agents, platform teams must evolve "Golden Paths" from human-centric workflows into structures that support autonomous execution patterns, such as low…

Monitor Azure Functions across every hosting plan with Datadog

Datadog has expanded its observability for Azure Functions by introducing a Serverless Compatibility Layer that enables telemetry collection across all hosting plans and runtimes. …