Span Level Trace Querying And Filtering Via Graphql

via “span-level trace querying and filtering via graphql”

Open-source LLM observability — tracing, evaluation, OpenTelemetry, span analysis.

Unique: Strawberry GraphQL schema specifically designed for LLM trace patterns (model names, token counts, retrieval metadata) rather than generic span attributes, with built-in support for RAG-specific filters like 'retrieval_source' and 'embedding_model'

vs others: More intuitive than raw SQL queries for non-database engineers, and more flexible than Jaeger's UI-only filtering for programmatic access

via “filtered trace search and analytics with custom view creation”

🪢 Open source LLM engineering platform: LLM Observability, metrics, evals, prompt management, playground, datasets. Integrates with OpenTelemetry, Langchain, OpenAI SDK, LiteLLM, and more. 🍊YC W23

Unique: Virtualized table rendering with complex filter combinations and saved views, enabling efficient exploration of 10k+ traces without performance degradation or manual query writing

vs others: Supports complex filter combinations (vs simple search in competitors), with virtualized rendering enabling 10k+ trace display vs competitors limiting to 1k-5k traces

via “trace querying and filtering via graphql api”

AI Observability & Evaluation

Unique: Uses Strawberry GraphQL framework with type-safe schema generation from Python dataclasses, enabling automatic schema validation and IDE autocomplete for query construction. Translates GraphQL queries directly to optimized SQL rather than loading full datasets into memory.

vs others: More flexible than REST APIs for complex filtering scenarios and more efficient than full-dataset retrieval; GraphQL schema is self-documenting and supports introspection for dynamic client generation.