From generic AI to task‑focused operational agents

Dynatrace ready‑made agents are purpose‑built capabilities that apply Dynatrace intelligence to specific, recurring operational tasks. Each agent focuses on a clearly defined problem, such as explaining why a service is slow, summarizing unusual behavior in an environment, or helping you understand what changed and why it matters. These agents are designed to take a question or a signal based on the exact data that is in your environment and organization and turn it into a useful answer you can act on.

Because Dynatrace agents are ready‑made, there is no need to define prompts, train models, or design behavior from scratch. Each agent already knows:

• What type of input to expect,
• Which Dynatrace signals and context it should use,
• And what output types are most useful for each addressed problem type.

All available ready-made Dynatrace agents can be found in Dynatrace Hub.

Trigger agent actions with Dynatrace Workflows and the Dynatrace MCP Server

Ready‑made agents can be triggered automatically as part of Dynatrace Workflows or available wherever you already work via the Dynatrace MCP Server.

Using agents in Dynatrace Workflows

Dynatrace Workflows lets you run agents in response to events or on a schedule. Instead of manually asking questions about potential problems and remediation steps, the workflow autonomously responds to changes in your environment.

For example, the Kubernetes Troubleshooting Agent runs nine parallel queries for data enrichment, and Dynatrace Intelligence turns all the information into a structured diagnosis. Customize the agents to your needs, including instructions for human approval steps and automated remediation.

The fastest way to get started is with Dynatrace ready-made agentic workflow templates, currently available in a preview release. Instead of building from scratch, you get proven automations that summarize issues, suggest remediation, and deliver insights directly to the tools your teams already use.

Power users can go further by building their own agentic workflows that combine Dynatrace Intelligence actions with any trigger, data source, or integration in Workflows. Use cases range from auto-scaling Kubernetes clusters based on Dynatrace Intelligence forecasts to generating query-cost-optimization recommendations for stakeholders, to virtually any other automation your environment requires.

Using agents through the Dynatrace MCP Server

The Dynatrace MCP Server makes the agents available outside the Dynatrace web UI, without requiring you to deploy or operate any additional infrastructure. You can connect Dynatrace to any MCP‑compatible client in minutes, with no server to install, host, or maintain.

Through the tools exposed by the MCP Server, you can use natural language to query data in Grail®, check system health, and get problem analyses and remediation recommendations. This brings Dynatrace directly into the tools you already use, such as your IDE, Claude Code and Cowork, Microsoft Copilot, Slack, or automation platforms like n8n. The Dynatrace MCP Server also powers integrations with systems like Azure SRE, AWS DevOps, GitHub Copilot, Atlassian Rovo Ops, Amazon Q, and others.

This means agents are no longer tied to a single interface. You can ask Dynatrace questions and get grounded, production‑ready answers wherever you work, using the same agents and intelligence that power Assist and workflows.

Dynatrace Assist: a simple way to test ready-made agents

The quickest way to use a ready‑made agent and see how it works before you start creating a workflow is with . Dynatrace Assist lets you ask questions about your environment using natural language, without switching tools or setting anything up.

A simple way to start is with a real problem you already have. For example, when a service becomes slow, open Assist and ask a question such as “Summarize the open problems and highlight those that need immediate attention.” Assist interprets the question, evaluates the environment you’re working in, and pulls together relevant data and context using Dynatrace Intelligence. Instead of manually navigating metrics, traces, logs, and dependencies, you get an explanation grounded in what is actually happening in your system.

Continuing your conversation with Assist, you can refine the question or follow suggested drill‑downs. Assist supports this as a single flow, helping you move from an initial question to deeper analysis and, where applicable, to next steps. You’re not configuring an agent or defining behavior. You’re simply asking a question and letting Dynatrace coordinate the right intelligence and ready‑made agents behind the scenes.

This makes Assist your lowest‑friction entry point for using Dynatrace agents. You get a concrete result quickly, using the same data and context you already rely on in your daily work.

What’s next?

If you haven’t already, open Dynatrace Playground, or your Dynatrace tenant, and ask Dynatrace Assist a question to see the ready-made agents in action.

The post Dynatrace AI agents begin working for you on day one, and are built to grow with you appeared first on Dynatrace news.

Executives are looking for successful ways to run their digital ecosystems with AI as cloud and AI adoption reach unprecedented complexity. Organizations are increasingly recognizing that agentic AI on its own can’t deliver the consistent, trustworthy outcomes they expect. With 65% of enterprises investing in AI‑driven monitoring and automation, leaders now need trustworthy AI‑powered observability to shift from human‑driven operations to human‑supervised, autonomous digital ecosystems.

Key executive insights

• Alongside the rapid adoption of agentic AI, Dynatrace is uniquely architected for powering real‑time autonomous operations across organizations’ digital systems while also integrating seamlessly into broader agentic ecosystems.
• Dynatrace – pioneer of large-scale AI-powered root cause analysis – established predictive operations and now further redefines observability, taking the next step from automation to autonomous action by auto-remediating, auto-preventing and auto-optimizing.
• Dynatrace takes the guesswork out of AI by optimizing the balance between deterministic AI, contextual analytics, and stochastic AI to drive precise agentic answers and reliable actions.
• Dynatrace Intelligence is an agentic operations system in the Dynatrace platform, driving autonomous actions through orchestrating ready-made Dynatrace agents as well as external ecosystem agents.
• AI engineering, AI operations, agentic SRE get enabled by Dynatrace with real-time production feedback-loops from trusted AI.

Minimizing hallucinations and avoiding large language model data processing limits

One of the biggest fears of executives who build agentic frameworks is that generative AI can hallucinate and push their agents off course. CTOs also prioritize ensuring agentic systems have instant access to high‑quality information, so multi‑step agent workflows can execute quickly and reliably.

Hallucinations aren’t minor errors – they can trigger wrong actions leading to outages, security risks, and financial exposure. To make it worse, in agentic processing inaccuracies can accumulate and amplify.

The Dynatrace AI approach reduces the risk of hallucinations by maximizing the use of deterministic AI in its agents, allowing Dynatrace to deliver responses based on real-time data rather than probabilistic guesses. Also, Dynatrace continuously observes, maintains, and enriches this context through real-time dependency graphs, real-world context, and high-performance data lakehouse analytics—fueling precise analytics and real-time answers across the entire digital services and business landscape.

This level of contextual precision is critical because large language models cannot directly process petabytes of heterogeneous observability data. Their context windows are limited, and performance often degrades as input approaches the maximum length.

Dumping large amounts of data into AI requests can reduce quality. Models have finite context windows and can underweight important details in very long prompts. Curating and structuring only the most relevant information generally yield better results than providing everything at once.

To overcome these constraints, it’s essential to rapidly distill vast amounts of data into short, high-quality context—this is where contextual analytics, dependency graphs, and the AI-optimized data lakehouse Grail become critical differentiators.

Real-time analytics with instant visualization of dependencies and interactions

Agentic AI depends not only on the model it runs on but on the quality, context, and immediacy of the data it receives—real‑time, fact‑based inputs are essential to keep pace with agentic decision‑making.

Grail, the Dynatrace data lakehouse, provides this foundation by unifying observability, security, and business data at an exabyte scale. Its schema‑on‑read approach removes indexing overhead and enables any‑question, any‑time analytics of Grail data. Grail processes metrics, logs, traces, user behavior, security events, and business signals alongside directed dependency graphs, delivering deep insights instantly through zero‑latency, always-hydrated storage.

Complementing this, Dynatrace Smartscape continuously refines a real-time dependency graph of real-world dependencies across business, teams, digital services, processes, infrastructure, risks, and more. By dynamically uncovering both vertical and horizontal dependencies, Smartscape enables teams to understand how systems, business processes, and organizational structures interact. The latest generation of Smartscape real-time dependency graph can now also be augmented with custom entities, such as business data types, ownership information, and other meta-data and ownership details, so teams immediately know who to notify for fast, fact-based remediation.

Together, Grail and Smartscape provide the technical foundation for Dynatrace Intelligence to let agentic AI act on facts, not guesses, ensuring that AI-powered decisions are accurate, scalable, and actionable, a pre-requisite for reliable autonomous operations.

Dynatrace Intelligence

Dynatrace Intelligence is an agentic operations system at the core of the Dynatrace platform that fuses deterministic AI with agentic AI to drive a new level of reliability across observability and autonomous operations. It provides a unified intelligence layer where humans define the goals, and AI executes them with precision—guided by policies, context, and guardrails.

At the foundation are deterministic agents that create a highly reliable operational core. The Root Cause Agent, powered by deterministic causal AI, delivers answers with far greater speed and precision than LLM‑only approaches. The Analytics Agent distills petabytes of Grail data lakehouse data into concise, contextual intelligence, while the Forecasting Agent scales predictive capabilities across the environment. An Operator Agent oversees, orchestrates, and coordinates agentic team efforts to ensure optimal outcomes. These foundational agents power every other agent operating on the platform.

Building on this foundation are ready‑made, domain‑specific agents designed to extend and augment the work of Development, SRE, and Security teams. This fusion of deterministic AI, agentic AI, and specialized domain agents enables Dynatrace Intelligence to detect anomalies, predict issues, identify root causes, run complementary investigations, and plan and execute corrective actions—ultimately enabling auto‑prevention, auto‑remediation, and auto‑optimization.

Dynatrace already observes customers’ digital services, end‑user experiences, and AI stacks automatically, making these domain agents immediately impactful. As examples, for developers, Dynatrace detects rising mobile‑app crashes, analyzes the code paths, and produces an immediate fix suggestion—turning what used to take hours into seconds. For security teams, Dynatrace continuously monitors emerging threats and instantly checks the environment for related vulnerabilities or indicators of compromise, helping teams respond proactively before attackers can act.

New Assist Agents simplify the adoption and everyday use of Dynatrace, while Agentic Workflows empower customers to build their own agents.

Dynatrace Intelligence is here not only to remediate symptoms—like a production infrastructure overload—but also to detect the underlying root cause and generate actionable plans to fix the source of the issue.

AI-powered observability from Dynatrace enables the next generation software delivery life cycle process with AI engineering, AI operations, agentic SRE, and AI business analytics, through real-time facts from production systems.

Already today, Dynatrace Intelligence agents collectively power a wide range of real‑world use cases, from mobile app crash inspection and front‑end error explanation to infrastructure optimization, Kubernetes operations, and security‑context insights. They also accelerate tasks like anomaly and log‑pattern analysis, vulnerability validation, timeseries analysis, and even dashboard creation, with extensible workflows that let teams expand these capabilities as their needs grow.

Dynatrace Intelligence also coordinates a bi-directional interaction with the agentic ecosystem like AWS Kiro, GitHub Copilot, ServiceNow, Azure SRE agent, Atlassian Rovo and many others – e.g. submitting tickets, invoking a coding agent, assessing the risk of a new deployment, adjusting infrastructure, informing business workflows. It also responds when other systems request, for example, incident details, performance insights, business impact analysis, or resilience risk assessments.

Along with Dynatrace Intelligence, Dynatrace leads customers on a journey toward fully autonomous operations.

Journey to fully autonomous operations

Each organization goes through their own maturity stages and pace on the journey to autonomous operation; for the majority it can look like this:

Automated – a stage when an organization’s digital system executes pre‑defined workflows and actions automatically (based on AI‑generated answers) to support both reactive and predictive operations. I can say that currently, many organizations are striving to get to this stage or are already in it.

Digital systems must be automatable and observable. To validate this, workflows must be testable:

• Can you automate it?
• Can you observe it?
• Can you understand its behavior in real time?

Organizations don’t need to automate every single workflow. Once it’s possible to confirm that key use cases are both automatable and observable, organizations can progress to the next maturity stage.

Supervised Autonomous – in this stage, AI generates execution‑ready action plans with clear reasoning and acts only after human oversight and approval. In a “crawl‑walk‑run” approach, organizations start with small, repetitive tasks that require agentic AI instead of hard-coded workflows.

Key principles in evaluating ability for supervised autonomous operations:

• Reliability: For reliable decisions and actions, rely as much as possible on deterministic AI and analytics, and leverage generative AI for common sense and learned expertise.
• Transparency: Allow people to set goals and guardrails for AI, validate reasoning, review knowledge graphs, improve real-time feedback loops, and assess explanations to build trust.
• Feedback loop: Agentic AI relies on accurate factual inputs to create actionable plans, as well as precise real-time feedback to investigate plan details, refine them and validate execution.

Once the digital system consistently performs with human‑like review discipline, the organization is ready to move toward fully autonomous operations.

Fully Autonomous – in the future, fully autonomous stage, Dynatrace Intelligence acts independently to fulfill business goals and autonomously operates a wide range of aspects in successfully delivering software that end-users expect, requesting human input only when necessary. As much as Dynatrace uses AI to observe other AI within the cloud- and AI-native services customers run, it also continuously observes itself—to self-optimize, ensure compliance, and provide insights that help people set the goals. People still play a crucial role: they review outcomes, adjust goals, and refine instructions. As a result, organizations deliver and operate software with higher resilience, happier customers, and lower cost.

The fusion of deterministic AI and agentic AI sets Dynatrace apart by providing a reliable agentic AI-powered observability. It is an AI that observes other AI and helps organizations to build more resilient applications and better customer experiences.

The post Dynatrace Intelligence at the core of autonomous operations appeared first on Dynatrace news.

In my frequent interactions with customers implementing agentic AI, the expectations of two key audiences—executives and developers—quickly become apparent.

Executives are actively exploring how to implement agentic AI, with a strong focus on unlocking significant productivity gains. They expect AI automation to free up engineering time, fix software automatically, prevent outages, and take over the majority of 80% of non-feature tasks.

Developers are rapidly adopting AI for convenience and efficiency in their day-to-day work; it’s becoming as essential to them as internet access. For example, GitHub Copilot usage among developers rose from 17% in 2023 to 45% in 2024. They want AI to bring context and suggest precise error repairs, generate tests automatically, auto-collect information to fix vulnerabilities, and recommend optimizations based on real production insights.

The market is embracing agentic AI with growing excitement. KPMG’s AI Pulse Survey, 68% of business leaders plan to invest between $50 million and $250 million in generative and agentic AI technologies this year alone, up from 45% in 2024. Enterprises see it as a strategic priority and as an enabler for smarter automation. While the potential is real, the requirements to make the use of agentic AI robust and secure need a solid foundation.

Context as foundation for reliable agentic AI

Imagine your car won’t start, and you ask an online car assistant for help. Most would start by asking you vague questions or suggesting generic fixes (“Try a new battery”) because they don’t understand or know the context of the problem. The next one might tell you: “Your engine is entirely broken. You need a new one.” Now, imagine instead you bring the car to an automotive expert who not only sees the reason for not starting but also instantly analyzes the entire build of your car down to the exact configuration of parts, how they interact, and even what parts were installed in what order. They don’t just know that the motor and screw exist: they know the screw holds the ignition coil to the engine block, and not the other way around.

This is how agentic AI works with Dynatrace. Agentic AI works like a team of experts who know your car inside out: every screw and why and how the vehicle was built. It’s not guessing but rather operating with architectural clarity, automatically pinpointing the root cause because it understands how everything is connected. With Davis® AI Root Cause Analysis, Dynatrace analyzes more than three million problems accurately and at scale every 24 hours, every day.

Instead of fumbling through 100,000 parts, it navigates a precise causal (say, the 50 services that actually influence the outcome) thanks to Dynatrace Smartscape. It doesn’t reach for every tool in the shed, but instead picks the right one for your specific digital system, every time.

So, similarly in IT: instead of general comments (“Your system seems slow, maybe scale your servers”), engineering teams get granular insights: “User slowdown originates from a failed API call in payment service, due to a misconfigured feature flag introduced in deployment of branch ‘calculation update in payment service’.” That’s how Dynatrace delivers context in action.

Today’s AI-powered automation in Dynatrace already shows agentic behavior

Dynatrace has long been operating at the intersection of data, intelligence, and automation. In fact, many capabilities typically associated with agentic AI, such as autonomous root cause detection, preventive operations, causal inference (which today has become causal AI), and self-healing production environments, have already been running across our platform for a decade.

Take this example: Dynatrace automatically detects a capacity issue, anticipates seasonal fluctuations, rates it by customer and business impact, and recalibrates the production environment across a customer’s hyperscaler setup, all end-to-end. It carries out full analytical and planning steps, creates reconfiguration plans, and only then notifies a human for final governance. This isn’t hypothetical: thousands of customers around the world are already leveraging our trusted causal and predictive AI in production workloads that run their businesses. And hundreds are taking the next step, adopting preventive operations by carefully adding generative AI to automatically draft remediation workflows, simulate outcomes, and enhance decision-making, shaping the future of intelligent automation.

Dynatrace AI capabilities flag and remediate problems, surface insights, and feed them into IDEs. This process triggers ticket creation to the responsible teams and aligns them around automatically planned actions, including learning from past incidents while incorporating real-time facts in context.

To further evolve from automation to autonomy, Dynatrace magnifies its capabilities with agentic AI and delivers three reliable agentic AI requirements through an architecture built for intelligent action.

Leveraging agentic AI for redefined observability with Dynatrace

The future of observability is being redefined by a powerful triad: Knowledge, Reasoning, and Actioning.

1. Knowledge. Dynatrace transforms contextual full-stack observability data into fact-based, real-time knowledge optimized for AI access. The Grail massive parallel processing data lakehouse is schema- and index-free, boosting AI agents with limitless query permutations. Grail works in tandem with Dynatrace Smartscape dynamic topology, an auto-discovered, continuously updated knowledge graph. This allows AI to deliver precise insights efficiently and at petabyte scale, eliminating the need for redundant queries (hence, also the increased cost) while maintaining full context and performance integrity.
2. Reasoning. Dynatrace unifies causal, predictive, and generative AI to power expert AI agents that optimize the blend of deterministic logic with probabilistic and stochastic models, to provide precision and fact-based trustworthy decision-making, while minimizing risks of hallucinations. This enables context-aware decisions with built-in enterprise-grade safety, compliance, and observability of AI itself, ensuring transparency and trust to not only achieve a capable AI, but also a reliable one.
3. Actioning. Dynatrace turns high-level objectives into intelligent, automated actions, where humans define the goals and AI determines the best way to achieve them, both reactively and proactively. With AutomationEngine, AppEngine, and OpenFeature, it remediates, optimizes, and even triggers systemic fixes, transforming observability into a strategic business enabler.

Last, but not least: AI is already powering production workloads across global enterprises, but not all AI is created equal. To deliver real value, it must be reliable, context-aware, and purpose-built for an organization’s digital environment. Dynatrace is engineered to meet those demands, magnified with agentic AI that answers organizations’ specific needs and business outcomes.

The post Shaping the Future: Autonomous Intelligence by Dynatrace appeared first on Dynatrace news.

The world has become software-defined, distributed, and complex, creating a widening gap between digital complexity and our ability to manage and govern the systems that run businesses and organizations. To close this gap, we reimagine how observability works. It is no longer enough to collect and analyze telemetry after the fact. Organizations need trusted, intelligent systems that turn real-time data into reliable knowledge, apply advanced AI to reason through that knowledge, and take action to optimize outcomes at every level of the business.

This is the foundation of the Dynatrace 3rd-generation platform. We’ve spent the past two decades shaping the observability market. Today, we are transforming it from a rear-view mirror into a real-time control system for the modern enterprise. Thousands of organizations are already using Dynatrace 3rd generation to turn data into decisions and decisions into action. The result is faster innovation and stronger business results across every layer of the business.

A new model built on knowledge, reasoning, and actioning

Dynatrace 3rd generation introduces a new standard for observability and automation based on three foundational capabilities:

• Knowledge: The Dynatrace platform turns petabytes of real-time data into a continuously updated, queryable knowledge graph. Powered by Grail and Smartscape, it provides trustworthy, fact-based insights with real-time context across dynamic environments.
• Reasoning: Causal, predictive, and generative AI models work together to derive intelligent decisions. These models are context-aware, transparent, and built for enterprise-grade safety and compliance.
• Actioning: Dynatrace enables users to define goals and let intelligent automation determine the best path forward, through innovations like AutomationEngine, AppEngine, and OpenFeature. This shifts operations from reactive remediation to preventive operations and continuous improvement.

Together, these capabilities form the foundation for autonomous intelligence. Dynatrace doesn’t just provide visibility; it enables systems to understand and act. By continuously converting real-time data into trustworthy insights, applying AI to reason through business and technical context, and triggering intelligent, goal-based actions, Dynatrace transforms observability into a real-time engine for automation and impact.

This is not about removing humans from the loop. It’s about empowering teams to define outcomes and rely on the system to carry out the best path forward. As with any leadership decision, autonomy depends on the quality of information and confidence in its context. The same principle applies to AI systems. Dynatrace 3rd generation gives organizations confidence, allowing them to scale decision-making with speed and trust.

Trusted knowledge, not just data

Legacy observability platforms focus on collecting telemetry data. But to support real-time decisions, teams need a trusted knowledge foundation. Dynatrace eliminates silos between metrics, traces, logs, events, user sessions, and security signals by unifying them in Grail, our schema-on-read, massively parallel data lakehouse.

For the first time, users can run any query at any time, with Grail supporting dramatically higher concurrency than traditional observability platforms. There’s no cold storage, no indexing, and no need for rehydration. This unlocks a goldmine of observability data and turns it into reliable, real-time answers.

That data is then contextualized in real time by Smartscape, our dynamic topology engine, and made instantly accessible to AI agents. The result is not just visibility, but deep, evolving system knowledge, providing machine-speed decisions no other platform can match.

AI that reasons with real-time context

Dynatrace has long set the standard for causal AI in observability. With the 3rd generation platform, we expand that foundation by combining causal AI with predictive and generative models. These AI types work together to support decisions at machine speed, with full context. Whether it’s automatically identifying the root cause of a service degradation, forecasting capacity needs, or evaluating how to improve online customer experiences, Dynatrace AI operates with the reliability and transparency required in enterprise environments.

Now, organizations can pursue modernization, transformation, and agentic AI initiatives with greater confidence. Dynatrace helps make AI accessible and actionable by reducing friction, delivering answers precisely when and where they are needed. Davis CoPilot enables natural language queries, workflow generation, and seamless integration into IDEs. It provides intelligent assistance at every step and supports a broad range of use cases across observability, security, and business operations with explainability, precision, and trust.

Automation that adapts to your goals

Traditional automation is limited by what is explicitly scripted. Dynatrace has taken a different approach. With the 3rd generation platform, you define high-level goals, and the platform determines the best way to achieve them. By grounding automation in real-time, high-quality data and precise causal analytics, Dynatrace ensures that actions are driven by accurate understanding, not assumptions.

This goal-based automation can resolve incidents, optimize performance, reduce cost, and even generate pull requests that improve code quality.

For example, preventive cloud operations allow site reliability engineers to move from firefighting to strategic orchestration. Instead of chasing alerts, SREs can focus on managing service-level objectives and improving business outcomes. Dynatrace handles the rest.

Built for the future of cloud and AI

The complexity of cloud-native architectures, Kubernetes deployments, and emerging agentic AI models are already testing the limits of traditional observability. Dynatrace 3rd generation is designed for the future.

By unifying telemetry, security data, and business context into a single real-time graph powered by Grail, Dynatrace provides the AI-powered intelligence required to operate modern systems with confidence. And by embedding automation throughout the platform, teams can scale faster than headcount, without sacrificing control or trust.

Turning observability into intelligent action

Organizations like TELUS and Air France-KLM are already seeing results: faster resolution, improved resiliency, and reduced downtime.

“By combining our Agentic AI initiatives with Dynatrace’s AI Observability capabilities, we’ve successfully optimized our development and operations workflows. We’re driving innovation and delivering measurable business impact while reducing downtime.”

– TELUS

“The AI and predictive capabilities from Dynatrace were a differentiator. We’re confident that any problem that arises can be dealt with quickly, dramatically reducing operational and revenue impact.”

– Air France-KLM

The combined impact of agentic AI initiatives and AI-powered observability extends beyond IT. These outcomes drive business performance, from greater availability and productivity to better customer experiences.

What this means for your organization

The Dynatrace 3rd-generation platform defines the path forward to a future where software can understand, reason, and act. It helps your organization move from reactive to proactive, from fragmented tools to unified intelligence, and from scripted automation to AI-driven operations.

Whether you are focused on cloud modernization, application security, cost optimization, or AI governance, Dynatrace provides a foundation to build and scale with confidence.

This is the evolution of observability. One built on context, driven by reasoning, and capable of taking action.

The post Dynatrace 3rd-generation platform: Built for the world of Autonomous Intelligence appeared first on Dynatrace news.