iterators, iterables, dunder methods, Agents Towards Production and more
with some more interesting news, articles, packages and projects
News
The PSF has paused our Grants Program
The PSF has temporarily paused its Grants Program after hitting the 2025 budget cap earlier than expected.
PyCon UK 2025 Schedule
Articles
2048: iterators and iterables
Iterable and iterator are not the same, though they’re often confused. In this article, Ned Batchelder explained the difference with a simple example and a great analogy: the pages of a book are an iterable; a bookmark is an iterator.
# Ned Batchelder
Don't call dunder methods
Dunder methods (like __init__, __str__, etc.) are low-level hooks used by the Python interpreter to enable certain behaviours. While they’re powerful, using them directly is often unnecessary and can lead to confusing code. In this article, Trey Hunner explained why it’s best to avoid them unless absolutely needed, with clear examples and shared alternatives.
# Trey Hunner
Why Python’s deepcopy Can Be So Slow (and How to Avoid It)
Python’s copy.deepcopy() is powerful but very slow and memory-heavy. If you're considering using it, check out this article by Saurabh Misra. He explained why it's often overkill and shared alternatives - like preferring shallow copies, copying only what's needed and other tips to make your code faster and more efficient.
# Saurabh Misra
Using Python Pandas to Clean and Analyze Scraped Data
If you work with data, you probably know that 95% of the job is cleaning and prepping it. In this article, Hassan Rehan highlighted the importance of data cleaning and showed how to use Pandas for processing, along with key techniques to clean and analyse scraped data effectively.
# Hassan Rehan
Agents Towards Production
Nir Diamant’s GitHub repository features 30+ in-depth tutorials for building AI agents. He covered everything from stateful workflows, vector memory and real-time web search APIs to Docker deployment, FastAPI endpoints, security guardrails, GPU scaling, browser automation, fine-tuning, multi-agent coordination, observability, evaluation and even UI development. A good resource for anyone building serious AI systems.
# Nir Diamant
Interesting Packages and Projects to Explore
dynaconf - The dynamic configurator for your Python Project
Pyro5 - Remote object communication library
Rio - Build modern Websites and Apps just with Python
PyCryptodome - Cryptographic library for Python
YTSage - Modern YouTube downloader with a clean PySide6 interface
Recent Noteworthy Package Releases
cuda-bindings 13.0.0 - Python bindings for CUDA
pydantic-ai 0.6.0 - Agent Framework / shim to use Pydantic with LLMs
torchvision 0.23.0 - image and video datasets and models for torch deep learning
curl-cffi 0.13.0 - libcurl ffi bindings for Python, with impersonation support
trl 0.21.0 - Train transformer language models with reinforcement learning
pyvista 0.46.0 - Easier Pythonic interface to VTK
protovalidate 0.14.0 - Protocol Buffer Validation for Python
mlflow 3.2.0 - An open source platform for the complete machine learning lifecycle
redis 6.3.0 - Python client for Redis database and key-value store
peft 0.17.0 - Parameter-Efficient Fine-Tuning (PEFT)
cartopy 0.25.0 - A Python library for cartographic visualizations with Matplotlib