Python Weekly (Issue 705 July 3 2025)

Author

Rahul Chaudhary
July 03, 2025

Welcome to issue 705 of Python Weekly. Let's get straight to the links this week.

Create How-to Videos in Seconds with AI

Stop wasting time on repetitive explanations. Guidde’s AI creates stunning video guides in seconds—11x faster.

  • Turn boring docs into visual masterpieces

  • Save hours with AI-powered automation

  • Share or embed your guide anywhere

How it works: Click capture on the browser extension, and Guidde auto-generates step-by-step video guides with visuals, voiceover, and a call to action.


Articles, Tutorials and Talks

The article explains how to read all data, including biometrics, from an NFC passport chip on Linux using the pypassport tool, detailing the process of reconstructing the required MRZ (Machine Readable Zone) password from passport details and providing Python code for both MRZ generation and chip reading

The guide explains how to build robust, production-grade logging in Python by moving beyond basic print statements to use the logging module’s loggers, handlers, formatters, and filters for structured, contextual, and queryable logs. It covers best practices like configuring loggers by module name, using YAML or dictConfig for centralized setup, and adopting JSON output for modern observability platforms, enabling effective debugging and operational insight in production environments

The video explores whether Python’s built-in dataclasses are still useful in 2025, given the rise of tools like Pydantic, FastAPI, and SQLAlchemy, concluding that most production code now relies on specialized data structures from these libraries rather than standard dataclasses. However, the presenter still finds dataclasses valuable for rapid prototyping and domain modeling, as they offer a simple and readable way to represent concepts without external dependencies

There are many approaches to speeding up Python code; applying multiple approaches can make your code even faster.

The article demonstrates how combining the new Python package manager uv with PEP 723 allows you to write self-contained Python scripts that declare and manage their own dependencies using inline metadata, making one-off scripting much easier and more portable. It showcases practical examples, such as running scripts with uv run and creating executable scripts that fetch YouTube transcripts, highlighting how this approach streamlines setup and execution for Python developers.

The article recommends always using kw_only=True in Python dataclasses to require all fields be set via keyword arguments, which makes your code more maintainable by allowing field reordering without breaking callers and enabling flexible subclassing. This practice is especially useful for library authors aiming for backward compatibility, but note that kw_only=True requires Python 3.10 or newer.

The article introduces inline-snapshot-django, a new Django package for snapshot testing SQL queries, allowing developers to automatically capture and compare query outputs in their tests. It aims to provide a more detailed and developer-friendly alternative to Django’s assertNumQueries(), making it easier to debug and maintain test suites for database interactions.

Optimizing Django performance? Use values() instead of only() when you only need specific fields, as it skips full model instantiation and avoids triggering deferred field lookups—boosting query efficiency. The article provides practical guidance on when to choose values() for lightweight data access.

A Pythonic way to ensure that your statically-defined dicts are complete, with full source code.

The video is a step-by-step tutorial on building a full-stack podcast discovery web app using only Python, leveraging the Reflex framework for both frontend and backend development. It demonstrates integrating user authentication with Clerk, database management with Postgres, and data validation with Pydantic, showing how modern Python tools can be combined to create interactive, production-ready web applications.

Learn how to build a fast, minimal HTTP server using asyncio.Protocol, complete with routing, parsing, and response handling from scratch!

The article explains how to train and finetune sparse embedding models using Sentence Transformers v5, highlighting their advantages in hybrid search scenarios, interpretability, and performance compared to dense models.

The article details how Google Cloud's Agent Development Kit (ADK) enables the creation of autonomous, multi-agent cybersecurity systems like CyberGuardian, which coordinate specialized agents for real-time threat detection, analysis, containment, and automated recovery.

Switching from pip to uv in Dockerized Flask or Django apps results in about a 10x faster dependency installation and introduces robust, lockfile-based dependency management, improving both speed and reliability. The process involves replacing requirements.txt with pyproject.toml, adjusting Dockerfiles to use uv commands, and configuring environment variables for efficient, reproducible builds without virtual environments


Interesting Projects, Tools, and Libraries

Automatically rewrites the websites into markdown optimised for viewing in the terminal.

This API provides programmatic access to the AlphaGenome model developed by Google DeepMind.

Python For Nonprofits demonstrates how to use Python to retrieve, analyze, visualize, and share nonprofit data.

Scan for secrets in dangling commits on GitHub using GH Archive data.

A Tree Search Library with Flexible API for LLM Inference-Time Scaling.

Genai-bench is a powerful benchmark tool designed for comprehensive token-level performance evaluation of large language model (LLM) serving systems.

Universal JSON Encoder/Decoder for Python.

PostgreSQL Row Level Security for Django.

OctopusDash is a modern, open-source Django admin dashboard designed with a beautiful UI, powerful filtering, and granular permission control — crafted for developers and teams seeking more flexibility, clarity, and extensibility beyond the default Django admin.

CLI tool for developing and profiling GPU kernels locally. Just write, test, and profile GPU code from your laptop.

A fully open source framework for creating RL training swarms over the internet.

Daytona is a Secure and Elastic Infrastructure for Running AI-Generated Code.


New Releases

Upcoming Events and Webinars

There will be following talks

  • Simulator/Agent for the "PyAlley" Board Game

  • AI in the Cloud: Modern Software Stacks & Cloud-Native Best Practices

There will be a talk, Dark Data: The Untapped Resource That Could Transform Industries.

There will be following talks

  • The architecture of a Python app

  • Handling crash reports with Bugsink

  • Deploying python apps with django-simple-deploy

There will be following talks

  • Data wrangling in a modern terminal

  • Right-Sized Scaling: Python APIs at Billions of Requests Without the Complexity


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.