Python Weekly (Issue 710 August 14 2025)

Author

Rahul Chaudhary
August 14, 2025

Welcome to issue 710 of Python Weekly. Let's get straight to the links this week.

Unmanaged AI = Unmanaged Risk. Shadow IT Could Be Spreading in Your Org

You wouldn’t allow unmanaged devices on your network, so why allow unmanaged AI into your meetings?

Shadow IT is becoming one of the biggest blind spots in cybersecurity.

Employees are adopting AI notetakers without oversight, creating ungoverned data trails that can include confidential conversations and sensitive IP.

Don't wait until it's too late.

This Shadow IT prevention guide from Fellow.ai gives Security and IT leaders a playbook to prevent shadow AI, reduce data exposure, and enforce safe AI adoption, without slowing down innovation.

It includes a checklist, policy templates, and internal comms examples you can use today.


Articles, Tutorials and Talks

Python’s standard library is one of its greatest strengths, but many developers only scratch the surface. This video explores 10 powerful and sometimes overlooked modules that can simplify your code, improve performance, and eliminate unnecessary dependencies. From dataclasses and pathlib to functools, graphlib, and heapq, you’ll see practical, real-world examples of how to use these tools effectively.

James Bennett reflects that Litestar, formerly known as Starlite, is a standout async-first, type-hint-driven Python web framework. For the past 18 months, it has been his go-to choice for new projects at work, and he believes more developers should discover this under-the-radar gem.

HRT created a custom fork of Python that uses lazy imports based on PEP 690, ensuring modules are only loaded when actually used. This optimization drastically reduced script startup times and notebook latency across their internal monorepo architecture.

This post guides developers through sending high-resolution image messages using RCS (Rich Communication Services) with Python and Messages API, enabling richer media delivery directly to users’ default messaging apps. The tutorial covers prerequisites like Python 3.6+, Vonage API credentials, and a registered RBM Agent, and includes full working code on GitHub. SPONSOR

You don't have to be a compiler engineer to understand how your code runs in a JIT build of CPython.

The story of how the Python Wheel Variant design was developed.

Python performance myths and fairy tales

Python performance engineer Antonio Cuni from the PyPy project challenges common misconceptions at EuroPython 2025, showing that memory management is often a bigger performance bottleneck than interpretation. He also explores how tools like Cython, Numba, and emerging projects like SPy can broaden the situations where Python can run at high performance

The article explains how pytest's parametrize feature can simplify and reduce duplication in writing tests by allowing a single test function to run multiple cases with different inputs and expected outputs. It uses a clear step-by-step example with a simple function to demonstrate how parametrize automatically runs the test for multiple data sets, making tests easier to write, read, and extend. This avoids repetition and improves test maintainability.

PyPI introduced stricter upload checks for Python wheel packages to prevent ZIP parser confusion attacks, now rejecting archives with mismatched or ambiguous ZIP headers, duplicate files, or RECORD metadata issues. After a six-month warning period, starting February 1, 2026, non-compliant wheels will be blocked entirely, strengthening ecosystem security with minimal impact on most projects.

No more Github links and no more sharing Jupyter tokens!

This post walks readers through building and containerizing a Flask app using Docker, Celery, and RabbitMQ, then shows how to deploy it seamlessly to Render while avoiding common pitfalls. He provides hands-on guidance for setting up local development, Docker Compose, and deployment configuration using render.yaml and CloudAMQP for background task wiring.

RevSys switched from embedding JSON-LD directly in Django templates to generating it with Django Ninja and Pydantic, resulting in cleaner templates, modular schema logic, and built-in validation for structured data. This refactoring improves SEO maintainability and aligns content with Schema.org standards for richer search results.

This article explains how to use drop-in replacements like NVIDIA cuDF, cuML, and cuGraph to dramatically speed up common Python data science workflows. It provides specific examples for accelerating popular libraries such as pandas, Polars, scikit-learn, and NetworkX on a GPU with minimal to no code changes.


Interesting Projects, Tools, and Libraries

Tilf (Tiny Elf) is a simple yet powerful pixel art editor built with PySide6. It’s designed for creating sprites, icons, and small 2D assets with essential drawing tools, live preview, undo/redo, and export options.

Talk to Your AI Agents from Anywhere!

LangDiff is a Python library that solves the hard problems of streaming structured LLM outputs to frontends.

Cross-platform lua GUI framework with built-in theme support.

Standardize FastAPI error handling with APIException. Custom error codes, fallback logging, and beautiful Swagger UI integration.

Graph-based live audio manipulation engine implemented in Python.

Extract and convert data from any document, images, pdfs, word doc, ppt or URL into multiple formats (Markdown, JSON, CSV, HTML) with intelligent content extraction and advanced OCR.

Open source lightweight embedded database engine for Python that supports NoSQL, SQL, and Graph data models.

Index Django Models with Elasticsearch or OpenSearch and query them with the ORM.

A research prototype of a human-centered web agent.

Open Source Digital Signage Solution for Raspberry Pi and PC.


Upcoming Events and Webinars

There will be following talks

  • Client-side Python with Pyodide

  • ELI5: typing.Generic

  • Understanding Data Types and File Formats

There will be following talks

  • Stop Overfeeding Your AI: A Practical Guide to Context Optimization

  • Mastering real-time anomaly detection

There will be following talks

  • Repeatable repeats with rpeat ® - DevOps for Science

  • Predicting Software Vulnerabilities: From Code Metrics to LLM-based Detection

There will be a talk, Beyond the ORM: from Postgres to OpenSearch.

There will be a talk, Setting up Retrieval Augmented Generation (RAG).

There will be a talk, How to Train Python.


Our Other Newsletters

Programmer Weekly - A free weekly newsletter for programmers.

Founder Weekly - A free weekly newsletter for entrepreneurs featuring best curated content, must read articles, how to guides, tips and tricks, resources, events and more.