Datadog announced new agentic AI monitoring and experimentation capabilities to give organizations end-to-end visibility, rigorous testing capabilities, and centralized governance of both in-house and third-party AI agents.
The new capabilities include AI Agent Monitoring, LLM Experiments and AI Agents Console.
Datadog is bringing observability best practices to the AI stack. Part of Datadog’s LLM Observability product, these new capabilities allow companies to monitor agentic systems, run structured LLM experiments, and evaluate usage patterns and the impact of both custom and third-party agents. This enables teams to deploy quickly and safely, accelerate iteration and improvements to their LLM applications, and prove impact.
“A recent study found only 25 percent of AI initiatives are currently delivering on their promised ROI—a troubling stat given the sheer volume of AI projects companies are pursuing globally,” said Yrieix Garnier, VP of Product at Datadog. “Today’s launches aim to help improve that number by providing accountability for companies pushing huge budgets toward AI projects. The addition of AI Agent Monitoring, LLM Experiments and AI Agents Console to our LLM Observability suite gives our customers the tools to understand, optimize and scale their AI investments.”
Now generally available, Datadog’s AI Agent Monitoring instantly maps each agent’s decision path–inputs, tool invocations, calls to other agents and outputs–in an interactive graph. Engineers can drill down into latency spikes, incorrect tool calls or unexpected behaviors like infinite agent loops, and correlate them with quality, security and cost metrics. This simplifies the debugging of complex, distributed and non-deterministic agent systems, resulting in optimized performance.
In preview, Datadog launched LLM Experiments to test and validate the impact of prompt changes, model swaps or application changes on the performance of LLM applications. The tool works by running and comparing experiments against datasets created from real production traces (input/output pairs) or uploaded by customers. This allows users to quantify improvements in response accuracy, throughput and cost—and guard against regressions.
Datadog unveiled AI Agents Console in preview, which allows organizations to establish and maintain visibility into in-house and third-party agent behavior, measure agent usage, impact and ROI, and proactively check for security and compliance risks.
The Latest
Artificial intelligence (AI) is core to observability practices, with some 41% of respondents reporting AI adoption as a core driver of observability, according to the State of Observability for Financial Services and Insurance report from New Relic ...
Application performance monitoring (APM) is a game of catching up — building dashboards, setting thresholds, tuning alerts, and manually correlating metrics to root causes. In the early days, this straightforward model worked as applications were simpler, stacks more predictable, and telemetry was manageable. Today, the landscape has shifted, and more assertive tools are needed ...
Cloud adoption has accelerated, but backup strategies haven't always kept pace. Many organizations continue to rely on backup strategies that were either lifted directly from on-prem environments or use cloud-native tools in limited, DR-focused ways ... Eon uncovered a handful of critical gaps regarding how organizations approach cloud backup. To capture these prevailing winds, we gathered insights from 150+ IT and cloud leaders at the recent Google Cloud Next conference, which we've compiled into the 2025 State of Cloud Data Backup ...
Private clouds are no longer playing catch-up, and public clouds are no longer the default as organizations recalibrate their cloud strategies, according to the Private Cloud Outlook 2025 report from Broadcom. More than half (53%) of survey respondents say private cloud is their top priority for deploying new workloads over the next three years, while 69% are considering workload repatriation from public to private cloud, with one-third having already done so ...
As organizations chase productivity gains from generative AI, teams are overwhelmingly focused on improving delivery speed (45%) over enhancing software quality (13%), according to the Quality Transformation Report from Tricentis ...
Back in March of this year ... MongoDB's stock price took a serious tumble ... In my opinion, it reflects a deeper structural issue in enterprise software economics altogether — vendor lock-in ...
In MEAN TIME TO INSIGHT Episode 15, Shamus McGillicuddy, VP of Research, Network Infrastructure and Operations, at EMA discusses Do-It-Yourself Network Automation ...
Zero-day vulnerabilities — security flaws that are exploited before developers even know they exist — pose one of the greatest risks to modern organizations. Recently, such vulnerabilities have been discovered in well-known VPN systems like Ivanti and Fortinet, highlighting just how outdated these legacy technologies have become in defending against fast-evolving cyber threats ... To protect digital assets and remote workers in today's environment, companies need more than patchwork solutions. They need architecture that is secure by design ...
Traditional observability requires users to leap across different platforms or tools for metrics, logs, or traces and related issues manually, which is very time-consuming, so as to reasonably ascertain the root cause. Observability 2.0 fixes this by unifying all telemetry data, logs, metrics, and traces into a single, context-rich pipeline that flows into one smart platform. But this is far from just having a bunch of additional data; this data is actionable, predictive, and tied to revenue realization ...
64% of enterprise networking teams use internally developed software or scripts for network automation, but 61% of those teams spend six or more hours per week debugging and maintaining them, according to From Scripts to Platforms: Why Homegrown Tools Dominate Network Automation and How Vendors Can Help, my latest EMA report ...