- Python Weekly
- Posts
- Python Weekly (Issue 758 August 13 2026)
Python Weekly (Issue 758 August 13 2026)
Rahul Chaudhary
August 13, 2026
In partnership with
Welcome to issue 758 of Python Weekly. Let's get straight to the links this week.
100+ Claude Code hacks to ship code 10X faster
Top engineers at Anthropic and OpenAI say AI now writes 100% of their code.
If you're not using AI, you're spending 40 hours doing what they do in 4.
These 100+ Claude Code hacks fix that and help you ship 10x faster.
Sign up for The Code and get:
100+ Claude Code hacks used by top engineers — free
The Code newsletter — learn the latest AI tools, tips, and skills to code faster with AI in 5 minutes a day
News
Django's Steering Council has accepted the Django Enhancement Proposal DEP 20 to move Django to an annual release cycle. From January 2028, Django will make one feature release a year, giving every feature release the LTS-level three years of support, and version numbers will carry the feature release year: Django 2028, then Django 2029, and so on.
Articles, Tutorials and Talks
In this post, the authors break down non-cryptocurrency zero-knowledge proofs (ZKPs) using graph theory and Python. By implementing Protocol 4 from Goldreich, Micali, and Wigderson, they show how a prover uses randomized color permutations, nonces, and cryptographic hashes to iteratively prove they hold a valid 3-coloring for a graph without revealing the actual solution to the verifier.
Two lines of Python could crash CPython 3.14 through 3.16 because SET_ADD assumed it was always operating on a real set, while PEP 749 made __conditional_annotations__ rebindable from Python code. The post explains how that assumption broke, why it caused a type-confusion crash, and why stable releases and the development branch needed different fixes.
From Thomas's 1979 majority voting, to ABD's linearizable register, to Cassandra's read repair: we provide a runnable Python tour of quorum replication, and show how ABD does not solve consensus.
Python raw strings still follow normal string-literal lexing rules, which leads to surprising cases such as raw strings being unable to end with a single backslash. The post highlights surprising f-string parsing rules, where embedded expressions can contain quotes, comments, and multiple lines because they're parsed as full Python expressions rather than simple string substitutions.
Outsiders can assess whether a foundation model was truly built from scratch by analyzing architecture configurations, tokenizer overlap, and weight embeddings using a reproducible fingerprinting pipeline. While architecture and tokenizer artifacts provide the strongest evidence, weight analysis has limitations and cannot cleanly distinguish continued pretraining from training from scratch.
DSPy’s new Flex module allows optimizers like GEPA to improve programs by rewriting both their instructions and underlying Python code. This lets the optimizer route easy cases to fast deterministic code while reserving model calls for ambiguity, significantly reducing cost and latency while improving accuracy.
The post shows how to prototype a data pipeline locally on 97 million Polymarket orderbook rows, then run the same LazyFrame queries on 16 billion rows using distributed execution. The pipeline pre-aggregates the raw data into small Parquet artifacts in S3 that power a responsive Plotly Dash dashboard without scanning the full dataset on each request.
Django 6.1 deprecates EMAIL_BACKEND and related email settings in favor of the new MAILERS configuration, ahead of their removal in Django 7.0. The article shows how to migrate existing email settings and update code to use mailer aliases and the new email APIs.
Traditional chatbot testing that checks only final responses misses many agent failures, so this article builds an evaluation framework that progresses from routing checks and task completion to full trajectory testing of tool use, conversations, and intermediate actions. It also shows how to combine LLM-as-a-judge evals, multi-turn testing, and telemetry to debug agent behavior and validate real-world deployments.
jsonfold is a streaming post-filter that compacts pretty-printed JSON by folding small arrays and objects onto single lines while preserving existing serializers and custom encoders. It processes JSON incrementally without reparsing the document, keeping additional memory usage bounded while producing more compact, human-readable output for large JSON files.
Numba now runs entirely in the browser through JupyterLite, compiling Python functions to WebAssembly with a new LLVM-based execution engine and delivering substantial performance gains without a server. The work also unlocks browser-native support for the broader Numba ecosystem, including PyMC, PyTensor, and other scientific computing libraries.
emojet is an emoji library for Python: it converts between emoji and their names, in both directions, plus the searching and lookup functions that go with that. It covers the core API of the emoji package, a library that been available for this job since 2014, using the same names and the same data. The difference is that emojet does the work in Rust, running 3.5x faster for conversion, 70 times faster for deconversion, and using about 40% less memory.
The Composite pattern lets you treat individual objects and entire object hierarchies through the same interface, making tree-like structures much easier to manage. Using a simple Python game engine, the video demonstrates how this pattern naturally applies to game scenes, UI frameworks, file systems, menus, and similar hierarchical designs.
Interesting Projects, Tools, and Libraries
Claude SEO is an open-source SEO analysis plugin for Claude Code. It runs 25 sub-skills and 18 specialist agents in parallel across technical SEO, content quality (E-E-A-T), Schema.org markup, AI search optimization (GEO), local SEO, e-commerce, and international SEO.
Graph-Native Infrastructure for Context and Accountable AI Systems.
The most comprehensive time-series feature extraction package in Python.
Agent Memory Guard is an OWASP Incubator Project that prevents AI agents from being weaponized through their own memory. It implements MITRE ATLAS mitigation AML.M0031 (Memory Hardening) to defend against context poisoning attacks (AML.T0080).
Circuit breaker for Python: sync + async in one class, sliding-window rate, slow-call detection, Polly-style resilience pipeline, type-safe API.
A lightweight server and framework for turn-based multiplayer games.
Centaur is frontier, agentic infrastructure that you own. Centaur is like Claude Tag, but open source and on steroids.
Let your AI agent use your passwords and API keys - without ever letting it see them.
Upcoming Events and Webinars
There will be following talks
GraphNN with PyTorch for prediction of molecular properties
all([]) == True and any([]) == False
There will be following talks
Exploring Execution Modalities
Docstrings as a Database — Building a Documentation Pipeline with Python & 11ty
There will be following talks
From Discovery To Transaction: Personalizing Marketplaces with Intent-Aware Ranking
Speeding up PyFixest's fixed effects demeaning with new algorithms and Rust
There will be a talk, Silicon Societies: What happens when millions of AI agents use the Internet alongside us?
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.