🧱 Laravel

The Filament v5 Beta is Here! (2 minute read)

Filament v5 beta has been released primarily to adopt Livewire v4. Though a major version, it enables seamless upgrades without altering Filament’s core. More exciting updates are anticipated soon.

Laravel 12.36 released

In the latest Laravel version, you can throttle concurrent HTTP requests in Http::pool and Http::batch, enforce same-origin redirects, manage Eloquent appends via setAppends/withoutAppends, and use Filesystem::allDirectories() to recursively list directory paths as strings.

🧰 RectorPHP and Laravel: The automated refactoring tool we should have been using years ago (4 minute read)

RectorPHP streamlines Laravel upgrades by automating refactoring tasks, reducing upgrade time by 40–60%. It applies framework rules, type hints, and modern PHP syntax, integrating smoothly into CI workflows.

​

💡 Tip of the Week

Wait for a specific time when a lock is unavailable. Use ⁠Cache::lock() with ⁠block() to efficiently manage concurrent access to resources.

Cache::lock(’resource_key’, 12)->block(8, function () {
    // Lock acquired for 12 seconds after waiting up to 8 seconds...
    // Perform critical operations here
});

​

🐘 PHP

Evolving PHP Streams for Async, Security, and Performance (4 minute read)

The PHP Foundation, backed by the Sovereign Tech Agency, is modernizing PHP Streams, enhancing performance, error handling, networking, and security. This update aims to improve I/O efficiency and introduce modern mechanisms.

🎨 Voting Open for PHP 8.5 Release Page Design Contest (GitHub issue)

Today is the final day to cast your vote in the PHP 8.5 Release Page Design Contest. The jury has shortlisted the top designs, and now it’s your turn to select the winner. The winner will be announced on November 4. There are some really good designs, so don’t miss the chance to support your favorite!

My wishlist for PHP in 2026 (6 minute read)

Reflecting on the PHP evolution, the author shares a wishlist for future features, including PHP editions for opt-in changes, interface default methods, generics, and structs, aiming to enhance flexibility and efficiency.

​

📦 Package of the Week

TOON

Token-Oriented Object Notation (TOON) is a token-efficient alternative to JSON for LLM prompts, cutting 30–60% tokens via tabular arrays and minimal syntax. Includes spec, TypeScript encoder/decoder and CLI, benchmarks, and multi-language ports (including PHP).

​

✨ Misc

🤖 Why do AI models use so many em-dashes? (8 minute read)

The article explores the prevalence of em-dashes in AI-generated text, attributing it to digitization of older literature with higher em-dash usage. It examines various theories but finds training data changes most convincing.

How much should you spend on AI tools in 2026? (5 minute read)

Engineering leaders are increasingly focused on AI tool budgets for 2026, with nearly half allocating 1-3% of their budgets to these tools. As AI costs rise, a multi-vendor approach is becoming common to access various functionalities, from code authoring to project planning. However, leaders face challenges in determining ROI, with 86% uncertain about which tools are most beneficial. Emphasizing measurement and data-driven evaluation is crucial for understanding AI’s impact and making informed financial decisions.

GPT-5-Codex is a better AI researcher than me (10 minute read)

An experiment in training AI models on a laptop led to a 1.8M param transformer, showcasing the potential of “vibe research” with AI tools like Codex. The process involved testing models like n-grams and transformers, highlighting the benefits of distilling transformers from n-gram models for improved coherence and storytelling.