Nir Gazit
Nir Gazit
Fixes #2753 - [ ] I have added tests that cover my changes. - [ ] If adding a new instrumentation or changing an existing one, I've added screenshots from...
### Discussed in https://github.com/traceloop/openllmetry/discussions/2174 Originally posted by **dinmukhamedm** October 18, 2024 It would be very useful for debugging purposes (e.g. with OpenAI support) to have a unique identifier to an...
1) I see that we are adding custom attributes to metrics, However, TRACELOOP_ENTITY_NAME (traceloop.entity.name) is a very high cardinality attribute - https://github.com/traceloop/openllmetry/pull/2154/files#diff-caff857eafc9121a585e40cdac1c0815d97f474cd3d0bf1a23755572e3b38105R473 - Most destinations would have limits on such...
**Description:** We're experiencing a regression with instrumentation in the recent LlamaIndex v0.8.4. The following example provided by LlamaIndex does not work with our current setup: [Agent Example](https://legacy.ts.llamaindex.ai/examples/agent). **Steps to Reproduce:**...
Title: Enable Automatic Instrumentation for LangChain in OpenLLMetry-JS ## Description: Currently, users of openllmetry-js are required to perform manual instrumentation when tracing LangChain applications. This is documented in our [JS...
### Description Traceloop would like to offer the donation of OpenLLMetry instrumentation packages to the OpenTelemetry project. OpenLLMetry is an open-source project that pioneered the use of OpenTelemetry to instrument...
### Discussed in https://github.com/traceloop/openllmetry/discussions/3348 Originally posted by **hudyweas** August 28, 2025 Hi! Is official OpenLLMetry support for the [Agno](https://github.com/agno-agi/agno) agent framework on your roadmap? It's one of the biggest frameworks...
## Summary This PR adds a comprehensive medical doctor Q&A example to the sample-app that demonstrates OpenLLMetry/Traceloop instrumentation with OpenAI's GPT models. ### Features Added - **Medical Q&A Application** (`medical_qa_example.py`):...
## Summary - Fixed tool calling serialization in streaming responses to follow OpenTelemetry semantic conventions - Previously multiple tool calls created separate completions (completion.0, completion.1, etc.) - Now properly grouped...