Kedro
Kedro is an open-source Python framework for building production-ready data pipelines, originally developed by QuantumBlack, a McKinsey company.
Profile
Kedro provides an open-source Python framework that helps data scientists and engineers build reproducible, maintainable, and modular data pipelines for production environments.
Kedro is an open-source Python framework for building production-ready data pipelines, originally developed by QuantumBlack, a McKinsey company. First released in 2019, Kedro provides a modular, opinionated structure for data scientists and engineers to create reproducible, maintainable, and version-controlled data workflows. The project is hosted on GitHub under the kedro-org organization, where it has accumulated over 10,900 stars and 1,000 forks as of June 2026.
The core team is managed through a Technical Steering Committee (TSC), with recent appointments including maintainers, committers, and advisors from QuantumBlack and the broader community. Kedro's product portfolio centers on the Kedro framework itself, which includes features for pipeline abstraction, data catalog management, parameterized configurations, and built-in support for testing and documentation. The framework integrates with popular data tools such as Ibis, anyLogistix, and various cloud platforms.
Kedro is positioned as a toolbox for data pipeline reproducibility, competing with frameworks like Apache Airflow, Prefect, and Dagster, but with a stronger emphasis on project structure and data science workflow conventions. The project does not disclose standalone revenue, funding rounds, or headcount, as it is maintained by QuantumBlack, which operates as a unit within McKinsey & Company. Recent activity on the GitHub repository shows ongoing development, including security scanning skills, HTTP server additions, and parameter validation fixes.
The community engages through a Slack workspace and a Linen-based forum for best practices discussions. Kedro's market position is as a specialized tool for data teams within larger organizations, particularly those already using Python and seeking a standardized approach to pipeline development. The project's trajectory is tied to QuantumBlack's broader consulting and technology offerings, with no public information about external investment or spin-out plans. As of mid-2026, Kedro continues to evolve with contributions from both QuantumBlack employees and external contributors, reflecting a steady but not explosive growth pattern typical of open-source tools backed by a consulting firm.
Who buys this
- Data science and machine learning teams in large enterprises
- Consulting firms and system integrators building data solutions for clients
- Organizations standardizing on Python for data pipeline development
- Teams migrating from ad-hoc scripts to structured, version-controlled workflows
- Companies in regulated industries requiring auditability and reproducibility of data processes
Strengths and what to watch
Strengths
- Strong integration with the Python data ecosystem, including Pandas, PySpark, and Ibis, allowing teams to adopt Kedro incrementally without replacing existing tools.
- Built-in project templating and modular pipeline design enforce reproducibility and reduce technical debt, which is a common pain point in data science projects.
- Active community and corporate backing from QuantumBlack (McKinsey) provide a level of governance and long-term maintenance that many open-source data tools lack.
Watch for
- Kedro's development and roadmap are controlled by QuantumBlack, a McKinsey subsidiary, which creates a single point of failure if McKinsey's priorities shift or if the project loses internal sponsorship.
- The framework's opinionated structure can be a barrier to adoption for teams that prefer lighter-weight or more flexible orchestration tools like Airflow or Prefect.
- No public information about revenue, funding, or user adoption metrics makes it difficult to assess the project's commercial viability or community health beyond GitHub stars and commit counts.
Key Information
- Industry
- AI Frameworks, Tools & Libraries
- Founded
- 1986
Frequently Asked Questions
What is Kedro and what does it do?
Kedro is an open-source Python framework for building production-ready data pipelines. It helps data scientists and engineers create reproducible, maintainable, and modular workflows. Originally developed by QuantumBlack, a McKinsey company, it provides a structured approach to pipeline development.
What are the main features of the Kedro framework?
Kedro offers pipeline abstraction, data catalog management, parameterized configurations, and built-in support for testing and documentation. It integrates with tools like Pandas, PySpark, and Ibis, and enforces modular design to reduce technical debt and improve reproducibility in data science projects.
How does Kedro compare to Apache Airflow or Prefect?
Kedro emphasizes project structure and data science workflow conventions more than Airflow or Prefect. While Airflow and Prefect focus on orchestration, Kedro provides an opinionated framework for building reproducible pipelines. It can complement these tools rather than replace them entirely.
Who typically uses Kedro for data pipelines?
Kedro is used by data science and machine learning teams in large enterprises, consulting firms building data solutions, and organizations standardizing on Python. It also suits teams migrating from ad-hoc scripts to structured workflows, especially in regulated industries needing auditability.
What are the strengths of using Kedro?
Kedro integrates well with the Python data ecosystem, including Pandas and PySpark, allowing incremental adoption. Its built-in templating and modular design enforce reproducibility and reduce technical debt. Corporate backing from QuantumBlack (McKinsey) ensures governance and long-term maintenance.
What should I watch out for when adopting Kedro?
Kedro's development is controlled by QuantumBlack, a McKinsey subsidiary, creating a risk if priorities shift. Its opinionated structure may deter teams preferring lighter tools like Airflow. Also, no public revenue or adoption metrics makes assessing commercial viability harder beyond GitHub activity.
Sources
- github.com — GitHub repository showing project activity, stars (10.9k), forks (1k), recent commits, and TSC role updates as of June 2026.
- kedro.org — Official website describing Kedro as a toolbox for production-ready data pipelines and providing documentation and getting-started guides.
- linen-slack.kedro.org — Community discussion on Linen Slack about best practices for data quality issues within Kedro, indicating active user engagement.
- kedro.org — Blog post about building scalable data pipelines with Kedro and Ibis, referencing a case study from Virgin Hyperloop One.