Python internals
Improving the extraction and lookup paths that support static dependency analysis at project scale.
About Deban
Deban contributes hands-on Python engineering to ArchUnitPython. His work sits close to the implementation details that turn source analysis into dependable test feedback, including performance-sensitive lookup behavior, data consistency, and documentation of the graph and report model.
A Python architecture library has to handle the flexibility of the language without becoming vague. Static analysis should be conservative about what it can prove, efficient over a growing project, and explicit when an import or selector cannot be interpreted safely.
Deban also brings experience with APIs, database tooling, and backend design. That makes the rules concrete: domain isolation, adapter ownership, package direction, and code organization are daily engineering concerns rather than abstract governance terms.
Focus
Improving the extraction and lookup paths that support static dependency analysis at project scale.
Keeping graph records, keys, and report output consistent so downstream checks receive trustworthy inputs.
Translating practical API, persistence, and domain boundaries into useful Python examples and documentation.
Selected project work
Optimized type-checking range lookup in ArchUnitPython with binary search.
Corrected duplicate-key behavior in the project data model.
Expanded README guidance for dependency graph reports and everyday usage.
Working principle
“Static analysis should be honest about the source, efficient over the project, and specific in its result.”
Contact Deban
Talk with Deban about architecture testing, the ArchUnit library family, or how executable guardrails can fit your engineering workflow.
lks.niessen@gmail.com