| TL;DR: In its first year, the OpenSearch Observability TAG brought together contributors from 10+ organizations to drive OTLP-native ingestion, Prometheus metrics support (shipped in OpenSearch 3.5), the OpenSearch Observability Stack, AI agent observability, and Dashboards-as-Code, all guided by community-reviewed RFCs and monthly open meetings. |
One year ago, OpenSearch launched the OpenSearch Observability Technical Advisory Group (TAG)—a community-driven group providing strategic direction and technical guidance for observability within the OpenSearch ecosystem. On September 2, 2025, the inaugural meeting brought together participants from AWS, SAP, Apple, and Hilti, setting out to shape the future of open-source observability together.
As the TAG celebrates its first anniversary, here’s what the group has accomplished, how the community made it happen, and how you can join for year two.
What is the Observability TAG?
Established under the Technical Steering Committee, the Observability TAG brings together subject matter experts from across the observability landscape to:
- Provide technical guidance for observability applications using OpenSearch—spanning logs, metrics, traces, profiles, ingestion pipelines, visualization, alerting, and anomaly detection
- Align OpenSearch with industry standards and external projects like OpenTelemetry, Prometheus, and Perses
- Establish best practices and foster community growth
The TAG meets monthly and all meetings are open to the public. Meetings include RFC reviews, community proposal discussions, and recommendations to the OpenSearch Technical Steering Committee (TSC).
A year of milestones
Over 10 meetings, the TAG has reviewed RFCs, guided architecture decisions, and brought diverse perspectives from organizations including AWS, Uber, SAP, Apple, CERN, Hilti, NetApp, Red Hat, Logz.io, Paessler, and IIFON. Here are the key themes and outcomes.
Cross-community collaboration
The TAG has served as a bridge to external communities:
- Perses (CNCF): The Perses project founder presented in November 2025, sparking exploration of OpenSearch plugin integration and dashboard component reuse. SAP has since been driving Perses integration forward.
- Kubernetes Event-Driven Autoscaling (KEDA): SAP contributed an OpenSearch scaler for KEDA, reviewed in the June 2026 TAG meeting.
- CNCF Observability TAG: The TAG aligned with broader CNCF observability guidelines and shared insights across communities.
- OpenTelemetry Demo SIG: TAG members actively contribute to the OpenSearch exporter in OpenTelemetry Collector Contrib and to the OpenTelemetry Demo, where OpenSearch serves as a logs storage backend. The OpenSearch Observability Stack builds on the Demo to showcase a full observability experience end to end.
OpenTelemetry and standards alignment
OpenTelemetry has been a thread through every TAG meeting—from the very first session discussing native OTLP ingestion paths and the OpenTelemetry Collector OpenSearch Exporter, through ongoing work aligning with OTel semantic conventions. The TAG has consistently championed standards-first approaches.
SAP’s Karsten Schnitter presented an OTLP plugin for direct ingestion to OpenSearch at OpenSearchCon Europe 2026, with TAG members volunteering as mentors and providing architectural feedback on the RFC. Karsten’s proof of concept—an OpenSearch OTLP receiver plugin—is now open for feedback.
Native metrics and Prometheus integration
One of the most impactful discussions came in October 2025, when Uber presented an RFC for a metrics time series analytics backend to build native time-series capabilities within OpenSearch. The TAG provided feedback on architecture, query performance, and Prometheus compatibility.
In parallel, the TAG reviewed and guided the new Discover Metrics experience—providing feedback on the explore experience, time series analysis and monitoring, multi-query support, label filtering, and PromQL integration in OpenSearch Dashboards. This work led to Prometheus metrics support launching in OpenSearch 3.5, with Prometheus alerting and AlertManager integration following shortly after.
The OpenSearch Observability Stack
A cornerstone achievement is the OpenSearch Observability Stack—a complete, open-source, production-ready observability platform. What started as a community effort to lower the barrier to entry evolved into a fully integrated stack built on OpenTelemetry, OpenSearch, and Prometheus, with its own dedicated site at observability.opensearch.org. Originally incubated as “ATLAS” (Agents, Tracking, Logging, Analytics Stack), it was reviewed during the TAG’s January 2026 meeting.
The Observability Stack offers one-command deployment, APM with service maps, distributed tracing, log analytics with PPL, metrics dashboards, alerting, and AI agent observability—all with zero vendor lock-in.
AI and agent observability
As AI agents emerged as a critical workload, the TAG prioritized this early. Beginning in November 2025, the TAG reviewed the AI Assistant Framework for OpenSearch Dashboards, discussed agent tracing architectures, and provided guidance on aligning with OpenTelemetry GenAI semantic conventions.
In subsequent meetings, the TAG reviewed agentic observability RFCs, discussed MCP server integration for the observability stack, and helped shape agent skills for log analytics and trace investigation. OpenSearch now offers first-class AI agent observability—execution graph visualization, token usage tracking, tool-call tracing, and LLM evaluation frameworks—with Agent Health.
Dashboards-as-code and visualization
The TAG reviewed the Dashboards-as-Code RFC, enabling YAML-based, GitOps-friendly dashboard configuration. The group also guided the Dashboard Variables proposal, providing feedback on cross-signal filtering, PromQL/PPL language support, and the relationship between variables and filters.
Operational improvements
In February 2026, the TAG established a GitHub project board for tracking issues and action items, implemented a triage process for each meeting, and contributed observability priorities to the 2026 OpenSearch Roadmap.
The community
What makes the TAG effective is the diversity of perspectives around the table. Over the past year, participants have joined from a wide range of organizations and backgrounds.
TAG members:
- Orcun Berkem, Principal Engineer Observability, AWS Open Source Observability
- Michael Froh, Software Engineer, Apple
- Yupeng Fu, Senior Principal Engineer, AI Search, Apple
- Shenoy Pratik Gurudatt, Software Dev Engineer, AWS OpenSearch
- Dotan Horovits, Senior Developer Advocate, AWS Open Source and OpenSearch Observability TAG Lead
- Jonah Kowall, Senior Vice President of Software and Design, Spacelift
- Karsten Schnitter, Senior Software Engineer and Development Expert, SAP BEI
- Mikhail Stepura, Senior Software Engineering Manager for OpenSearch, Apple
- Dr. Jürgen Walter, Engineering Manager, SAP
- Shuyi Zhang, Software Engineering Manager, Uber
- Anirudha (Ani) Jadhav, Senior Manager Software Development, AWS OpenSearch
Regular participants and contributors from Apple, SAP, CERN, Hilti, NetApp Instaclustr, Red Hat, Logz.io, IIFON, and the Perses project—plus many individual contributors who attend, ask questions, and shape the conversation.
This year, the TAG promoted its first active contributor to a TAG member: Anirudha (Ani) Jadhav. The road is open for anyone to grow with the community and expand their influence.
Thank you to everyone who has participated, reviewed RFCs, opened issues, and contributed code. This TAG belongs to the entire community.
Get involved
The Observability TAG is open to everyone—formal membership is not required to participate or contribute. Here’s how to get started:
- Join the conversation on Slack — Connect on OpenSearch Slack in the #observability channel
- Attend meetings — TAG meetings are public and held monthly. Check the TAG page for the schedule and meeting link
- Review the charter — Understand the scope and goals in the governance charter
- Participate in RFCs — Browse open issues on GitHub and leave feedback on proposals
Looking ahead
The observability landscape continues to evolve rapidly. Areas where the TAG will focus in year two include:
- Evolution of AI and agent observability
- Deeper Prometheus and metrics ecosystem integration
- Integration with cloud-native projects and the broader ecosystem
- Further alignment with OpenTelemetry semantic conventions
- Improving ease of use and documentation for observability deployments
- Data transformation and ingestion pipeline enhancements
Open-source observability thrives when the community comes together to build in the open. The OpenSearch Observability TAG has accomplished much in year one—and this is just the beginning.
Join the TAG and help shape the future of open-source observability.