Beyond the Benchmark: Why the “Best” Python Dependency Manager Might Not Be Best for You
summary
Benchmarks from the LinkedIn Loop repo show UV leads benchmarks with Poetry second and pip trailing across cold and warm installs and updates; tooling/install times matter because CI spins up often. GitHub analysis of ~22M Python repos finds requirements.txt ~560K, Poetry ~40K, and UV ~14K — combined modern-tool adoption <5% — so migration costs and human/ROI factors outweigh raw speed. A local MCP server is built that pulls official docs, uses Rust bindings for fuzzy search, and provides AI-assisted, tool-backed comparisons; demo mostly works but needs a stable server toggle.
AI-written summary · about
chapters (13)
- 0:00Introduction
- 1:27Python dependency managers
- 2:52Dependency management complexity
- 3:33Speaker's journey
- 5:03Benchmarking package managers
- 9:00Benchmark reliability
- 10:31Adoption rates
- 14:58Ecosystems driving Python deps
- 17:28Costs beyond performance
- 21:29Community fragmentation
- 22:35AI in dependency management
- 23:18Demo: MCP server
- 27:01Q&A
related talks
33muv without sunblock: faster than ever dependency and project management
26mDependency Management in Python
52mManaging Dependencies in Python with Poetry
20mHow I got unstuck with Python
51mTread Lightly When Building and Testing Your Python Project
31mBuilding an extremely fast Python package manager, in Rust
42mSo much depends upon... a python app's dependencies
07mLightning Talk – Python Dependencies the Right Way