Episode

#524: 38 things Python developers should learn in 2025

Podcast: Talk Python To Me
Published: Oct 20, 2025
Duration seconds: 4155
Processing state: processed
Canonical source: https://talkpython.fm/episodes/show/524/38-things-python-developers-should-learn-in-2025
Audio: https://talkpython.fm/episodes/download/524/38-things-python-developers-should-learn-in-2025.mp3?v=2
JSON: /v1/public/podcasts/talk-python-to-me/episodes/524-38-things-python-developers-should-learn-in-2025
Markdown: /podcast/talk-python-to-me/524-38-things-python-developers-should-learn-in-2025.md

Actions

POST https://stenobird.com/v1/public/podcasts/talk-python-to-me/episodes/524-38-things-python-developers-should-learn-in-2025/transcription-requests
Idempotently request low-priority transcript generation for this episode.
GET https://stenobird.com/podcast/talk-python-to-me/524-38-things-python-developers-should-learn-in-2025.md
Read the agent-friendly Markdown representation of this episode resource.

Summary

Python development is undergoing a fundamental shift with the arrival of free-threaded CPython and high-performance tooling. This episode explores the essential technologies for 2025, from modern packaging with uv to edge computing with PyScript.

Topics

Python
CPython
Docker
Kubernetes
uv
DuckDB
WebAssembly
Data Science
AI Workflows
Software Engineering

Highlights

Main idea: The removal of the GIL in CPython is enabling true parallel execution in Python threads
Practical takeaway: Use 'uv' for lightning-fast package management and 'tilt' to simplify Kubernetes development workflows
Practical takeaway: Leverage Docker Compose and containers to maintain isolated, reproducible development environments
Failure mode: Avoid complex Kubernetes setups for simple web apps; use lightweight alternatives like K3s or stick to Compose when possible
Main idea: The rise of DuckDB and WebAssembly (Wasm) is bringing powerful data processing and AI capabilities directly to the browser

Chapters

11:20 The End of the GIL: A discussion on the CPython core team's move toward supporting free-threaded Python and the implications for parallel execution.
27:10 Containerized Development: How Docker Compose and Tilt can be used to manage complex local stacks and ensure environment parity.
32:40 Kubernetes for Web Developers: Evaluating the learning curve of Kubernetes versus Docker Compose and when to adopt orchestration tools.
38:05 Modern Packaging with uv: Exploring how the uv tool is revolutionizing Python installation speeds and environment isolation.
59:00 AI and Edge Computing: The impact of MCP, PyScript, and WebAssembly on running AI models and data processing at the edge.
1:04:00 High-Performance Data with DuckDB: Using DuckDB and Arrow for efficient, in-process analytical workloads and browser-based data science.