Vector Search Using Ollama for Retrieval-Augmented Generation (RAG)

LangChain Python Tutorial: 2026’s Complete Guide

If you’ve read the blog post How to Build Chatbots With LangChain, you may want to know more about LangChain. This blog post will dive deeper into what LangChain offers and guide you through a few more real-world use cases.

5 Essential Itertools for Data Science

Learn 5 essential itertools methods to eliminate manual feature engineering waste. Replace nested loops with systematic functions for interactions, polynomial features, and categorical combinations.

Vector Search with FAISS: Approximate Nearest Neighbor (ANN) Explained

ScrapeGraph - Natural Language Web Scraping

Web scraping without selector maintenance. ScrapeGraphAI uses LLMs to extract data from any site using plain English prompts and Pydantic schemas.

Sorting Strategies for Optional Fields in Django

How to control NULL value placement when sorting Django QuerySets using F() expressions.

Text Classification With Python 3.14’s zstd Module

There is commonality between text classifiers and compression and there are algorithms out there to do one with the other, but it requires an incremental compressor. Python 3.14 added zstd which supports this feature, allowing Max to take a stab at doing ML with a compressor.

Claude Code from the beach: My remote coding setup with mosh, tmux and ntfy

Django (anti)patterns

A website and repo with 39 common antipatterns, listing them as well as suggested changes. Worth a look!

Why using [n] on a Django QuerySet can be unsafe?

Indexing a QuerySet can return nondeterministic rows because slicing does not add ordering, unlike first, which orders by primary key.

Django: profile memory usage with Memray

ㅤ

From Python 3.3 to today: ending 15 years of subprocess polling

Django: profile memory usage with Memray

Use Memray to profile Django startup, identify heavy imports like numpy, and reduce memory by deferring, lazy importing, or replacing dependencies.

Asyncio Is Neither Fast Nor Slow

There are many misconceptions on asyncio, as such there are many misleading benchmarks out there. This article looks at how to analyse a benchmark result and to come up with more relevant conclusions.

Nothing to Declare: From NaN to None via null

Explore the key differences between NaN, null, and None in numerical data handling using Python. While all signal “no meaningful value,” they behave differently. Learn about the difference and how to correctly handle the data using Pydantic models and JSON serialization.

The Hidden Cost of Python Dictionaries

Learn why Python dicts cause silent bugs and how NamedTuple, dataclass, and Pydantic catch errors earlier with better error messages.

I Created a Game Engine for Django?

Multiplayer Snake implemented in Django using Django LiveView, 270 lines of Python, server side game state, WebSocket driven HTML updates, no custom JavaScript.

Golfing APL/K in 90 Lines of Python

Fun With Mypy: Reifying Runtime Relations on Types

This post describes how to implement a safer version of typing.cast which guarantees a cast type is also an appropriate sub-type.

How to Switch to ty From Mypy

The folks at Astral have created a type checker known as “ty”. This post describes how to move from Mypy to ty, including in your GitHub Actions.