SigNoz vs Windmill
Side-by-side trajectory, velocity, and editorial themes.
SigNoz pairs an AI teammate with enterprise access control and wide cloud coverage
SigNoz, the open-source ClickHouse-backed observability platform, is advancing on three fronts at once. Noz, its AI teammate that answers plain-English questions across live telemetry, is now general to all cloud users. Cloud and integration coverage keeps widening — Azure services and six new onboarding sources including PlanetScale and Cloudflare Workers — while fine-grained, role-based access control entered beta for Cloud and Enterprise. Underneath, Query Builder v5, trace-detail rework, and a ClickHouse version bump continue.
The platform is maturing from a query tool into an investigation surface: an AI layer to drive analysis, RBAC and self-service API keys to make that safe in larger orgs, and out-of-the-box integrations to shorten onboarding. Notably, the access-control work is explicitly framed around feeding read-only keys to the SigNoz MCP Server for AI tooling, tying the enterprise and AI tracks together. Expect Noz and MCP access to keep converging with the permissions model.
Next likely moves: Noz gaining more write-style actions beyond suggestions, RBAC graduating from beta with role assignment delegated, and continued ClickHouse-version-gated features like JSON trace attributes.
Windmill is quietly turning its orchestrator into a DuckLake-native data platform.
Windmill remains a script-and-workflow orchestrator, but its recent output has narrowed almost entirely to the data layer: DuckLake materialization, partitioned assets, freshness, and now schema contracts. The changelog reads less like a general automation tool and more like a managed lakehouse being assembled on top of the existing job engine, with most of the heavier machinery gated to Enterprise Edition.
The direction is a governed, self-maintaining data platform: fork-scoped data environments for isolated iteration, freshness SLAs with an auto-healing watchdog, scheduled compaction and retention, range backfills, and save-time schema validation between producers and consumers. Each release hardens a different operational corner of the DuckLake stack rather than adding breadth, so Windmill is deepening into data-ops rather than widening its surface.
Schema contracts shipped as non-blocking warnings, so the next likely move is to tighten that loop — enforceable contracts or richer lineage on the DuckLake graph. Continued Enterprise gating of the data-ops features (maintenance, freshness watchdog, range backfill) also points to a productized data-platform tier.
See more alternatives to SigNoz →
See more alternatives to Windmill →