Agentic systems for what comes next.
We are a small team of veteran builders led by Wes McKinney.
Kenn is building a new stack of development and knowledge systems for
the agentic era.
projects
- roborev Continuous code review and quality assurance for coding agents v0.63.0 1,583
- msgvault Offline archive and search system for email, texts, and chat, with agent-friendly CLI and MCP interfaces
v0.18.0 1,958
- agentsview Unified archive, search, and token intelligence for over 20 coding agents
v0.39.0 4,588
- middleman Local-first rapid development workflow system for modern coding platforms like GitHub and GitLab
143
- kata Lightweight local-first issue tracker for coding agents, with human-friendly terminal UI
v0.12.1 322
- docbank Local-first system of record for documents (PDFs, images, text files, etc.) for you and your agents
v0.11.0 45
team
-
Wes McKinney
Influential open-source data systems developer, entrepreneur, and angel investor, focusing on data processing, developer tools, and community building. Wes started the Python pandas and Ibis projects and co-created Apache Arrow. Once upon a time an AI skeptic, he is now "all in" on agentic engineering and AI-powered application development.
-
Phillip Cloud
Software engineer who works across systems programming, developer tooling, and analytics. Phillip has solved hard problems in digital video, Rust, computer vision, and Python-native CUDA APIs. He was for many years a prolific contributor to open-source analytics tools like pandas, Arrow, and Ibis. These days he spends much of his time building with AI.
-
Matt Jacobs
Software engineer with deep roots in distributed data systems. Matt was a tech lead at Google Cloud, working on cloud database products. He worked on Apache Impala at Cloudera as a contributor to the core database and a PMC member. He cares about the hard technical problems underlying our architectures, spanning resource management, reliability, and scalable infrastructure. These days he's excited by the opportunities presented by agentic engineering, and using new AI capabilities in ways that teams can make high-quality software to tackle hard problems.
-
Marius van Niekerk
Experienced craftsman working at the intersection of AI, software, and infrastructure. Marius has built data systems across ad tech, health tech, backend data architecture, and open-source infrastructure, usually close to the places where reliability and scale start to matter. Across conda and conda-forge, he has helped to steer the ecosystem and to turn messy packaging problems into tools developers rely on every day.