Topics /
The most popular blog posts about software engineering on Hacker News. 419 posts.
We’re happy to officially launch torchao, a PyTorch native library that makes models faster and smaller by leveraging low bit dtypes, quantization and sparsity. torchao is an accessible toolkit of techniques written...
Early C had no separate operators for & and && or | and || . (Got that?) Instead it used the notion (inherited from B and BCPL ) of "truth-value context": where a Boolean value was expected, after "if" and "while" and so forth, the & and | operators were interpreted as && and || are now; in ordinary expressions, the…
From the Zed Blog: In this episode of Zed Decoded, Thorsten and Antonio walk through the performance optimizations Antonio when working on Zed's Rope data structure before they then pair on adding more optimizations
Follow the first article of a series in which we'll try to summarize our journey with Astro, sharing many cool little tricks and details we found out (or totally came up with) in the process.
Safari 17.4 adds vertical writing mode support for form control elements across macOS, iOS, iPadOS, and visionOS.
Every year for Black Friday Cyber Monday (BFCM), we put together a real-time visualization of purchases made through Shopify-powered merchants worldwide. This year we're cooking up something big, and in anticipation we wanted to show you how we built the globe last year. Video of BFCM 2023 Globe Besides going for…
Posted by Lars Bergstrom – Director, Android Platform Tools & Libraries and Chair of the Rust Foundation Board Back in 2021, we announced...
Unsloth fixing Google's open-source language model Gemma.
This blog post summarises a new garbage collection algorithm called Mark-Scavenge, which highlights how using reachability as a proxy for liveness in moving GCs leads to unnecessary data movement and how we can address this. This work is from the latest paper within the research collaboration between Oracle and…
Datomic and Content Addressable Techniques: An Ultimate Data Wonderland
“The What and Why of Adapter based Parameter Efficient Fine Tuning: Understanding Its Purpose and Significance”
How we cut build times by 50% and shipped a solution for scale.
GDB loses significant functionality when debugging binaries that lack debugging symbols (also known as “stripped binaries”). Function and variable names become meaningless addresses; setting breakpoints requires tracking down relevant function addresses from an external source; and printing out structured values…
OpenAI’s GPT-4 Turbo with Vision model scores worse on aider’s code editing benchmarks than all the previous GPT-4 models. In particular, it seems much more prone to “lazy coding” than the existing GPT-4 Turbo “preview” models.
S3ctor is a core developer of OpenMW, and has, like most of us to be fair, odd interests. This is the incredible story how he mixed two immensely cool things: Quake, and Morrowind, narrated by our …
The Vulkan Working Group at Khronos has developed a set of video format decode and encode extensions, collectively referred to as Vulkan Video. Today, with the release of Vulkan 1.3.277, the Working Group is proud to announce the new Decode AV1 video extension. Similar to the Vulkan project in its goals, AV1 is a…
In September, machine learning engineers at Mozilla filed a bug report indicating that Firefox was consuming excessive memory and CPU resources while running Microsoft’s ONNX Runtime (a machine learning library) compiled to WebAssembly.
How to kill off a top source of browser exploits
Manage entire Python projects and even Python itself with a single unified tool.
Blog for DoltHub, a website hosting databases made with Dolt, an open-source version-controlled SQL database with Git-like semantics.
OpenTelemetry round for 2023 and what to expect in 2024
Can solving Super Mario Bros. help solve your distributed systems issues?
fsspec is secretly everywhere, and boss
When we received feedback our Notebooks UI was taking too long too load, our engineers dove into ways to improve the developer experience — bringing some load times from 30 seconds down to less than one.
SST now natively supports building containerized applications.
This report performs the first public analysis of MMTLS, the main network protocol used by WeChat, an app with over one billion users. The report finds that MMTLS is a modified version of TLS, however some of the modifications have introduced cryptographic weaknesses.
This essay explores the connection between Erlang's process supervision, state management and durable execution.
Enable JavaDoc documentation comments to be written in Markdown rather than solely in a mixture of HTML and JavaDoc @ -tags.
How we spent the last 3 years restructuring the AntennaPod code base.
