Building an OpenTelemetry Normalizer for GenAI Observability: Lessons from groundcover (Part 1)
By
Anais Dotis
Fresh out the oven, still warm. Top of the tray.
Summary
This article shares lessons from groundcover engineers building a GenAI SDK, framework, and provider-agnostic AI observability solution using OpenTelemetry (OTel). It highlights the challenge that while many claim OTel support, few emit consistent attributes. The piece emphasizes groundcover's philosophy that observability data collection should be straightforward and unbounded, allowing SREs to focus on incident response and root cause analysis rather than instrumentation and data processing. It's the first part of a series on normalizing OTel for GenAI workloads.
Key quotes
· 4 pulledEveryone says they support OpenTelemetry. However, nobody actually emits the same attributes.
One of groundcover's main values is that o11y data collection should be straightforward and unbounded.
We believe that SRE's should focus on addressing incidents and root cause analysis instead of instrumentation, data processing, data gaps, dashboarding, and materialized views.
We aim to make AI Observability with OTel as simple as it currently is to collect
You might also wanna read
PromptLayer: AI observability tool for tracing multi-step workflows, costs, and failures
PromptLayer is an AI observability tool for developers that provides a unified timeline and waterfall view to trace AI requests, workflows,
Latitude: An Observability and Quality Platform for AI Agents
Latitude is an observability and quality platform for AI agents, designed to help developers detect and resolve failure modes before they re
Kōan: A tool for live-streaming AI agent reasoning, tool calls, and decisions
Kōan is a tool that provides live streaming visibility into AI agent operations, showing reasoning, tool calls, decisions, and errors in rea
Why small pull request policies can backfire on software quality
The article critiques a common software engineering policy that limits pull requests (PRs) to small sizes (e.g., 500 lines, few files). Whil
apenwarr.ca·2h agoHow Anthropic contains Claude's expanding access across its products
Anthropic describes how it has evolved its approach to granting Claude, its AI assistant, increasingly broad access to internal systems over
Testing Cursor's Jira integration: How ticket quality affects AI agent performance
Cursor launched a Jira integration that lets developers assign tickets directly to an AI agent, eliminating context switching. The author te
bit.ly·3h ago