/Deep Dive

Reverse Engineering Linear's Sync Engine

- Evan Hu tl;dr: “I am not affiliated with the Linear team, nor have I consulted them while writing this article. As a result, there may be inaccuracies or discrepancies with the actual implementation. However, I've made every effort—especially by watching relevant talks and comparing LSE to well-studied operational transformation (OT) approaches—to ensure that my description of the LSE approach is as accurate as possible. I hope it serves as a valuable reference for building a similar collaborative engine.”

featured in #623

The AI Engineering Stack

- Gergely Orosz Chip Huyen tl;dr: The authors discuss: (1) Three layers of the AI stack. Application development, model development, infrastructure. (2) AI engineering versus ML engineering. Similarities and differences. (3) Application development in AI engineering. The three main focus areas: evaluation, prompt engineering, and AI interfaces. (4) AI Engineering versus full-stack engineering.

featured in #619

Code Quality > Quantity

tl;dr: Explore BlueOptima’s deep dive into code quality and its real impact on maintainability. The report reveals how static metrics provide a clear picture of technical debt and genuine cost-saving opportunities, and why consistent refactoring and proven design patterns are key to sustainable software performance.

featured in #599

XOR

- Simon Tatham tl;dr: “Recently I was called on to explain the ‘XOR’ operator to somebody who vaguely knew of its existence, but didn’t have a good intuition for what it was useful for and how it behaved. For me, this was one of those ‘future shock’ moments when you realise the world has moved on. When I got started in computers, you had to do low-level bit twiddling to get anything very interesting done, so you pretty much couldn’t avoid learning about XOR. But these days, to a high-level programmer, it’s much more of an optional thing, and you can perfectly well not know much about it.”

featured in #593

Taking A Look At Compression Algorithms

- Moncef Abboud tl;dr: “What is compression? We represent data using bytes (1 byte = 8 bits). The more bits we have, the more storage space we need and the longer the transmission time. If we can represent some data with fewer bits, we can save on both storage and time it takes to process or transmit. So compression translates to cost savings and better performance. And at scale, we are talking about millions of $.”

featured in #583

No GPS Required: Our App Can Now Locate Underground Trains

tl;dr: “Thanks to our clever engineering, we can now predict your location in a subway tunnel using your phone’s vibration signature.” This post dives into how. 

featured in #568

How Passwordless Authentication Works - A Deep Dive

- James Walker tl;dr: Passwordless authentication is a better option than traditional passwords, improving security and increasing convenience by removing the challenges associated with passwords. James explains how passwordless works using methods like magic links, biometrics, or one-time codes, and the benefits it provides engineering leaders and CTOs.

featured in #544

How Passwordless Authentication Works - A Deep Dive

- James Walker tl;dr: Passwordless authentication is a better option than traditional passwords, improving security and increasing convenience by removing the challenges associated with passwords. James explains how passwordless works using methods like magic links, biometrics, or one-time codes, and the benefits it provides engineering leaders and CTOs.

featured in #535

How Apple Accidentally Broke My Spotify Client

- Roberto Frenna tl;dr: “I turned on my work laptop and started catching up on everything that happened during my absence. Music makes this process more enjoyable, so I started up Spotify and started playing some tunes. I was very surprised (and mildly pissed) when playback abruptly stopped and Spotify seemingly went offline.” Roberto dives deep to understand the issue. 

featured in #483

Abracadabra: How Does Shazam Work?

- Cameron MacLeod tl;dr: Shazam does the following to register a song: it calculates a spectrogram of a son, extracts peaks from that spectrogram, pairs those peaks up into hashes and stores the collection of hashes for a song as a fingerprint. Cameron discusses these in depth, as well as how Shazam recognizes an audio sample and matches it against its database.

featured in #472