Engineering

Cross-posted from X / @_can1357 In fact only the edit tool changed. That’s it. 0x0: The Wrong Question The conversation right now is almost entirely about …

(See how I cleverly did not mention AI in the title!) You know we have seen more than our fair share of slop reports sent to the curl project so it seems only fair that I also write something about the state of AI when we get to enjoy some positive aspects of this technology. … Continue reading A new breed of analyzers →

Learn how to apply established UI patterns for a more organized and maintainable codebase and discover the benefits of layering architecture in React development

Even if you’ve been doing JavaScript for a while, you might be surprised to learn that setTimeout(0) is not really setTimeout(0). Instead, it could run 4 milliseconds later: Nearly a decade a…

Picture this: You are a dev in a payment services company. The thing with payment transactions is that you are supposed to process each transaction (e.g. sen...

What is the type of 2 + 2 = 4?

A comprehensive, practical guide to understanding traces and spans in OpenTelemetry—how they work, how they relate, and how to instrument real Node.js / TypeScript applications effectively for deep insight and faster debugging.

A personal tech blog by Stephen Funk

In a world where some frameworks declare the VDOM “pure overhead,” we’ve made a radical bet: we’ve doubled down on the VDOM and moved it entirely off the main thread. This isn’t a defense of the…

A reverse proxy is a critical piece of software commonly found in various setups within a distributed system. You may have encountered it as a proxy enabling service mesh, a load balancer…

In Part 1 of this series, we explored a high-level overview of reverse proxies and dived deep into connection management. This post shifts our focus to the intricate world of HTTP handling within a reverse proxy.

The main idea is, instead of injecting what you need, you ask for it using an effect.

Node.js runs JavaScript code in a single thread, which means your code can do one task at a time and can’t use multiple cores. To remedy this, Node.js introduced the worker-threads module, so that you can create threads and execute multiple JavaScript tasks in parallel on multiple cores. In this tutorial, you’ll create a Node.js app with a CPU-intensive task. You’ll use the worker-threads module to offload the CPU-intensive task to another thread to avoid blocking the main thread.

Nullsafe is a new static analysis tool that is used at Meta to detect NullPointerException (NPE) errors in Java code.

See how you can easily add syntax-highlighted code blocks to your diagrams.

Great software architects don’t just make technical decisions—they balance competing priorities, navigate messy org charts, and align engineering with business strategy. This post explores the habits and mindset that separate competent architects from truly impactful ones, and why those differences matter. Whether you're eyeing the role or already in it, these are the practices that move the needle.

Explore popular caching strategies for improved application performance, including Cache Aside, Read/Write Through, Write Back, and more

This is my personal notes on monads, that some might find helpful. I’ve included more information in the appendix if needed. In my experience it is often more useful to talk about what a specific…

Explained as simply as possible… but not simpler.

Since 2002, I have been keeping track of all the tricky bugs I have come across. Nine years ago, I wrote a blog post with the lessons learned from the bugs up till then. Now I have reviewed all the…

Paul Khuong's personal blog. Some Lisp, some optimisation, mathematical or computer.

This post explores the enduring challenge in software programming of how to pass invisible contextual information, such as loggers or request contexts, through applications without cumbersome explicit parameter passing. It examines various approaches throughout history, including dynamic scoping, aspect-oriented programming (AOP), context variables, monads, and effect systems. Each method offers a unique solution, from the simplicity of dynamic scoping in early Lisp to the modularity of AOP and the type-safe encoding of effects in modern functional programming. The post highlights the trade-offs of each approach, such as the unpredictability of dynamic scoping or the complexity of monad transformers. It also touches on how context variables are used in modern asynchronous and parallel programming, as well as in UI frameworks like React. The author concludes by noting that the art of passing the invisible is an eternal theme in software programming, and this post provides valuable insights into the evolution and future directions of this critical aspect of software architecture.

Many years ago I played League of Legends for about 2-3 years. If you have ever played it as well, you know that there is a champion called Caitlyn. Caitlyn is straightforward to play but hard to master. You need to use her abilities aggressively early in the game otherwise mid game you will have a hard time. I would compare coding in React with playing Caitlyn. It's simple, every developer who knows a bit of Javascript can write some components, connect them and voila - there is your web app.

How to test your service for different browsers and devices: user needs, verified browsers, adapting to change.