OllyGarden Extends AI Agent to Identify Instrumentation Gaps

OllyGarden has enhanced its AI agent, Rose, to not only optimize the collection of telemetry data but also to identify and address gaps in existing instrumentation. After securing an additional $4 million in funding, the company's founder, Juraci Paixão Kröhling, explained that the newly added Minimum Viable Instrumentation (MVI) capability to the Rose AI agent enables it to pinpoint the most vital sources of telemetry data that should be instrumented using OpenTelemetry, an open-source framework for data collection supported by the Cloud Native Computing Foundation (CNCF).

This extension builds upon Rose's initial purpose of helping DevOps teams minimize the telemetry data they need to collect and store by identifying the most relevant logs, traces, and metrics. OllyGarden's goal is to reduce the overwhelming amount of telemetry data DevOps teams must handle, which varies but is increasingly growing as more AI workloads are deployed.

The company's AI agent continuously assesses the quality of collected telemetry data, detects issues, and proposes source-level fixes for DevOps teams to implement, thereby preventing irrelevant data from reaching observability platforms. Over the past year, OllyGarden claims to have assisted DevOps teams in identifying potential reductions in telemetry data volume by up to 85% across their logs.

While observability platforms will likely play a significant role in enabling DevOps teams to integrate AI agents into their workflows, the challenge lies in reducing the volume of data exposed to these agents to ensure they deliver high-quality outcomes at the lowest possible cost. DevOps teams may still need to supervise AI agents in the near future as fully autonomous deployment in production environments is yet to be widely adopted.