Dynatrace For Ai

Dynatrace is an observability and security platform, and its AI Observability capability extends that platform to generative AI applications, large language models and agents. It gives teams end to end visibility across the whole AI stack, from the user experience layer through orchestration frameworks and retrieval pipelines down to the model and the GPUs and TPUs beneath it, so failures, cost spikes and quality problems can be traced to a root cause. It is built for engineering, operations and platform teams running AI features in production, especially in larger organizations that already need application and infrastructure monitoring. AI Observability tracks token usage, latency, cost and invocation errors per model, monitors RAG pipelines and vector databases, and observes agent tool usage and agent to agent communication. Guardrail monitoring detects hallucinations, prompt injection, PII leakage and toxicity, and continuous evaluation uses a model as judge to score response quality and alert when it drifts. Full prompt to response lineage with long term storage supports audit and compliance reporting. It sits alongside the wider platform, including APM, log analytics, digital experience monitoring, application security and the Grail data lakehouse. Dynatrace sells the platform through one annual commitment that every capability draws down at published rate card prices. Examples from the rate card are Foundation and Discovery at $7 per host per month, Infrastructure Monitoring at $29 per host per month, full stack application and infrastructure monitoring at $58 per 8 GiB host per month, and Kubernetes Platform Monitoring at $1.40 per pod per month. A free trial is available. Dynatrace has announced its intention to acquire Arize, an AI observability and evaluation company, which signals further investment in this area.