Observability startup Dash0 announced this week that it acquired Berlin-based continuous profiling specialist Polar Signals. The deal adds continuous profiling to SignalStore, Dash0’s OpenTelemetry-native data platform.

Continuous profiling has been part of the developer toolbox for years and provides teams with an ongoing view of where running applications spend CPU time, allocate memory and consume other resources. Many observability platforms now include some form of it. Polar Signals offers a less common capability: continuous profiling of Nvidia CUDA workloads in production. Dash0 said the technology can profile both AI training and inference workloads, with visibility down to individual GPU kernels, to help monitor resource consumption and pinpoint where performance problems are occurring.

Dash0 will integrate Polar Signals’ profiling data into SignalStore and feed it into Agent0, its AI agent for production operations. Agent0 can already correlate telemetry with source code and prepare pull requests for fixes. Profiling adds another source of runtime evidence to those investigations, which Agent0 can use when deciding what changes to propose. Dash0 will also let external coding agents access that profiling data through its MCP server, giving them runtime context for suggesting changes.

The company is also developing AutoTune, an Agent0 capability that will analyze profiling data on a schedule and look for opportunities to improve CPU and memory use. When it identifies an optimization, AutoTune is designed to open a pull request with the proposed change and the profiling evidence behind it. Dash0 said the system will not merge those changes on its own. Developers will still need to review the changes and decide whether to accept them.

The acquisition grew out of a conversation between Dash0 CEO Mirko Novakovic and Polar Signals founder Frederic Branczyk at a dinner in Berlin. Branczyk wrote that Novakovic described the cost, performance and operational challenges Dash0 was encountering with ClickHouse, while Polar Signals had already built several iterations of a database for similar workloads. As they talked, Branczyk said, it became clear the companies had been “working toward the same thing from opposite ends.”

That overlap suggests why the deal extends beyond continuous profiling. Dash0 is also acquiring Polar Signals’ Great Lakes storage engine and engineering team. Polar Signals built Great Lakes to handle the high-cardinality data generated by profiling, and Branczyk said Dash0 eventually plans to replace ClickHouse beneath SignalStore with Great Lakes. The goal is to store metrics, logs, traces and profiles in the same engine, he said. The deal also includes Parca, Polar Signals’ open source continuous profiling project, which the company said will remain open source and continue to be maintained.

Financial terms of the deal were not disclosed. Dash0, founded in 2023, raised $110 million in March at a $1 billion valuation and said it would use some of the funding to pursue strategic acquisitions. The company now has more than 750 customers.

The acquisition also complements Dash0’s OpenTelemetry-based architecture. OpenTelemetry Profiles, the project’s profiling specification, entered public alpha in March, extending the OpenTelemetry data model to include profiles alongside traces, metrics and logs. Polar Signals has contributed to that work, bringing Dash0 a team already involved in shaping OpenTelemetry’s emerging profiling standard.