Dagger

Dagger develops a platform for automating software delivery: building, testing and shipping codebases repeatably and at scale. Its core is an execution engine and cross-language system API for orchestrating containers, filesystems, secrets, git repositories and network tunnels, paired with SDKs for eight languages (Go, Python, TypeScript, PHP, Java, .NET, Elixir and Rust), an interactive REPL, and an ecosystem of reusable modules. Pipelines are defined in general-purpose programming languages rather than shell scripts or proprietary YAML, and run in containers with sandboxed orchestration functions, explicit and strictly typed host dependencies, just-in-time construction of intermediate artifacts, and content-addressed incremental caching.

The product is positioned as local-first: once configured, the same automation runs on a laptop, an AI sandbox, a CI server or dedicated cloud infrastructure, with a Linux container runtime (e.g. Docker, or Docker Desktop on macOS/Windows) as the only dependency. Observability is built in — every operation emits OpenTelemetry spans with logs and metrics, viewable in a terminal TUI, a web view, or exported to backends such as Jaeger or Honeycomb. Documentation describes typical uses including linting and formatting, end-to-end tests with real services, running code generators, interactive debugging of broken builds via drop-in terminals, and agents running Dagger checks inside their development loop.

Alongside the open-source engine (Apache-2.0 licensed), the company offers Dagger Cloud, described as a control plane that complements the engine with workflow visualization (Traces) and operational insights aggregated across an organization's engines in both development and CI contexts. Dagger Cloud is a "bring your own compute" service rather than a hosting service, and works with any CI environment where the engine can run, including GitLab, CircleCI, GitHub Actions, Jenkins and Tekton. The company collectively describes the Dagger Engine, Dagger Cloud and the SDKs as the "Dagger Platform" and frames it as a "devops operating system."

Founding story

Dagger was founded by Solomon Hykes, Sam Alba and Andrea Luzzardi, who identify themselves as the creators of Docker, with the stated aim of bringing order to the complexity of software deployment. Their public launch materials frame the problem as devops engineers assembling delivery pipelines from throwaway scripts and CI-specific configuration, causing dev/CI drift, CI lock-in and low reuse between pipelines.

Business model

Open-core: the Dagger engine and SDKs are distributed as an open-source project under the Apache-2.0 license and installable via a CLI (for example, Homebrew), while Dagger Cloud provides a commercial hosted control plane layer offering trace-based workflow visualization and operational insights on top of customer-provided compute.

Sources describe a free open-source engine plus Dagger Cloud, a "bring your own compute" hosted service; specific pricing or revenue terms are not disclosed in the available material.

Traction

The open-source repository has approximately 16,200 stars, 913 forks, 229 watchers and 12,454 commits. The project shipped a public beta in March 2022 and has progressed through multiple release lines to a 1.0.0-beta.10 documentation version. Users are reported to run Dagger with CI providers including GitLab, CircleCI, GitHub Actions, Jenkins and Tekton.

Latest developments

Current documentation describes Dagger as a CI orchestration engine centered on workspaces and installable modules (for example eslint, prettier, playwright and Go code generators) invoked through commands such as `dagger check`, `dagger install`, `dagger generate` and `dagger api call`, along with guidance on upgrading to workspaces. A 1.0.0-beta.10 version line is documented, and the feature set now emphasizes agent-friendly workflows in which AI agents run sandboxed, cached Dagger checks.