Open Source by McKinsey is an ecosystem of projects we're excited to share with the world. We recognize that open source has become a foundational element of delivering organizational transformation for our clients globally. It is also our way to contribute to inclusive and sustainable growth and to narrow the digital divide.

Our open-source work is maintained by Tech & AI at McKinsey.

We believe in the power of open source and are excited to contribute to the community. We invite you to explore our ecosystem, join in the development, and propose new initiatives! Together, we can drive open innovation.

Here on the Open Source by McKinsey ecosystem, you'll find the following open-source projects:

• Vizro - a low-code toolkit for building high-quality data visualization apps. Use it to create customized dashboards in minutes without the need for advanced coding or design experience.

• Agentic runtime for Kubernetes (ARK) - a declarative, Kubernetes-native framework for building portable, scalable, and provider-agnostic agentic applications. By defining what agents should do rather than how they do it, Ark eliminates vendor lock-in and ensures your applications stay adaptable as AI evolves.

• Agents at Scale Marketplace - a curated collection of ready-to-deploy services, agents, MCP servers, executors, Helm charts, and DevSpace configurations for the Agents at Scale ecosystem.

• QuantumBlack Design System - a shadcn-based component library that provides accessible components built with Base UI and Radix UI primitives, and styled with design tokens. Add them to your project as source files through the shadcn registry, not as an NPM package.

Tech & AI by McKinsey maintain additional open source projects, including:

• Kedro - a toolbox for production-ready data science. It uses software engineering best practices to help you create data engineering and data science pipelines that are reproducible, maintainable, and modular. Kedro was donated to the Linux Foundation's AI & Data Incubator to become an open standard in machine-learning engineering.
• MLRun - an open MLOps platform for quickly building and managing continuous ML applications across their lifecycle. MLRun integrates into your development and CI/CD environment and automates the delivery of production data, ML pipelines, and online applications, significantly reducing engineering efforts, time to production, and computation resources.
• Nuclio - a high-performance "serverless" framework focused on data, I/O, and compute-intensive workloads. It integrates with popular data science tools, such as Jupyter and Kubeflow, supports a variety of data and streaming sources, and can be executed over CPUs and GPUs.