Alexander Reelsen

@spinscale
296 Followers
379 Following
1.5K Posts
Husband, dad, enjoys working distributed, hacking #java #crystallang #concurrency #kotlin #web #serverless, Basketball/Streetball fan
Websitehttps://spinscale.de
GitHubhttps://github.com/spinscale
LocationEmsdetten, Germany
Twitterhttps://twitter.com/spinscale
Alexander Reelsen18h ago

The Reviewer Isn't the Bottleneck

Not sure yet I agree on everything in there, but that would be boring. "The constraint has moved to context, review quality, and accountability. The parts that don’t compress as easily."

https://rishi.baldawa.com/posts/review-isnt-the-bottleneck/

The Reviewer Isn't the Bottleneck

AI tools are flooding PR queues and the instinct everywhere is to call review the bottleneck. I think that’s the wrong question. The reviewer is the last sync point before production changes. The goal shouldn’t be how to remove the gate, but how to make it cheaper to operate.

Rishi Baldawa
Alexander Reelsen1d ago

vmprint - a javascript typesetting engine with PDF output

Uses a JSON format to create exact PDFs, also has a markdown to AST converter on its website. That could remove the need for headless chrome for many projects, especially on the command line.

https://github.com/cosmiciron/vmprint

GitHub - cosmiciron/vmprint: A pure-JS, tiny typesetting engine with bit-perfect PDF output on everything—from Cloudflare Workers to the browser. No more Headless Chrome to just print text. AND -- try the static demos on the home page below and see for yourself.

A pure-JS, tiny typesetting engine with bit-perfect PDF output on everything—from Cloudflare Workers to the browser. No more Headless Chrome to just print text. AND -- try the static demos on the h...

GitHub
Alexander Reelsen2d ago

emdash - an opensource agentic development environment

ADE seems to be the hottest thing at the moment, with jetbrains launching Air. I guess that IDEs have a high interest, that users stay in their ecosystems while writing much less code. So far I haven't found an ADE that I like. Anyone has CLI/TUI tools to recommend?

https://github.com/generalaction/emdash

GitHub - generalaction/emdash: Emdash is the Open-Source Agentic Development Environment (🧡 YC W26). Run multiple coding agents in parallel. Use any provider.

Emdash is the Open-Source Agentic Development Environment (🧡 YC W26). Run multiple coding agents in parallel. Use any provider. - generalaction/emdash

GitHub
Alexander Reelsen3d ago

GenDB: An LLM-Powered Generative Query Engine

uses LLM agents to generate instance-optimized query execution code, tailored to your specific data, faster than Umbra and DuckDB in the shown queries. Plus a linked paper.

https://github.com/SolidLao/GenDB

GitHub - SolidLao/GenDB: GenDB, an LLM-Powered Generative Query Engine Built for the Future

GenDB, an LLM-Powered Generative Query Engine Built for the Future - SolidLao/GenDB

GitHub
Alexander Reelsen4d ago

deff - an interactive, side-by-side file review for git diffs with per-file navigation, vertical and horizontal scrolling, syntax highlighting

Using the github UI I always find it hard to review big PRs exceeding several thousand lines of code. Maybe a different viewer is helpful. Taking recommendations here.

https://github.com/flamestro/deff

GitHub - flamestro/deff: interactive, side-by-side file review for git diffs with per-file navigation, vertical and horizontal scrolling, syntax highlighting, and added/deleted line tinting

interactive, side-by-side file review for git diffs with per-file navigation, vertical and horizontal scrolling, syntax highlighting, and added/deleted line tinting - flamestro/deff

GitHub
Alexander Reelsen5d ago

mdvi - cli based Markdown viewer

Written in Rust, uses ratatui, vim keybindings. I am pretty happy with bat as my main markdown viewer at the moment due to it's simplicity, but maybe I am not working with enough markdown.

https://github.com/taf2/mdvi

Alexander Reelsen6d ago

Nobody Gets Promoted for Simplicity

"Anyone can add complexity. It takes experience and confidence to leave it out."

https://terriblesoftware.org/2026/03/03/nobody-gets-promoted-for-simplicity/

Nobody Gets Promoted for Simplicity

We reward complexity and ignore simplicity. In interviews, design reviews, and promotions. Here’s how to fix it.

Terrible Software
Alexander ReelsenMar 12

The inner workings of TCP zero-copy

Great post, what is required for zero copy to make it work on the sending and receiving side of the Linux kernel and that it's more of a throughput and not a latency feature plus io_uring support.

https://blog.tohojo.dk/2026/02/the-inner-workings-of-tcp-zero-copy.html

The inner workings of TCP zero-copy

/dk/tohojo/blog
Alexander ReelsenMar 11

About memory pressure, lock contention, and Data-oriented Design

Fantastic long blog post about digging into a performance issue step-by-step on the matrix Rust SDK.

https://mnt.io/articles/about-memory-pressure-lock-contention-and-data-oriented-design/

About memory pressure, lock contention, and Data-oriented Design

I illustrate how _Data-oriented Design_ helped to remove annoying memory pressure and lock contention in multiple sorters used in the Matrix Rust SDK. It has improved the execution by 98.7% (53ms to 676µs) and the throughput by 7718.5% (from 18K elem/s to 1.4M elem/s)! I will talk about how the different memories work, how we want to make the CPU caches happy, and how we can workaround locks when they are a performance bottleneck.

mnt.io
Alexander ReelsenMar 10

Bf-Tree is a modern read-write-optimized concurrent larger-than-memory range index in Rust from MS Research

SSD optimized (requires random writes as fast as sequential writes), for small records, concurrent, larger than memory. Haven't read the linked paper yet.

https://github.com/microsoft/bf-tree

GitHub - microsoft/bf-tree: Bf-Tree is a modern read-write-optimized concurrent larger-than-memory range index in Rust from MS Research.

Bf-Tree is a modern read-write-optimized concurrent larger-than-memory range index in Rust from MS Research. - microsoft/bf-tree

GitHub