System Architecture

Secure programming for the AI era

How a Slack shout-out, a dusted-off academic theory, and a spaghetti monster led an AWS team to crack an elusive code—and deliver greater reliability and performance for customers.

A modern, COBOL-inspired language for the JVM. Inspired by COBOL — not COBOL.

An agent shows up at your API. It has a verifiable identity, a clear purpose, and a list of scopes it would like. Your existing API management platform canno...

A high-level introduction to db9.ai: serverless Postgres, files, branching, and agent-native workflows in one backend.

Someone calls you at work and says, “Don’t tell anyone, but…” If you are like most people, there are one or two people you will pass it along to with the same admonishment. …

The universal .discovery layer for AI agents. DNS-AID has 2 repositories available. Follow their code on GitHub.

This blog post tells the story about how a nightly entity sync evolved from microservices and queues into a scalable batch architecture —…

By Parth Jain, Rakesh Sukumar, Yingwu Zhao, Renzo Sanchez & Nathan Fisher How we built a living map of our distributed infrastructure to…

What if the filesystem was backed by a real database?

And what it reveals about the hard tradeoffs in multi-tenant execution platforms

I came across the concept of “Nitro Enclaves” while reading through the AWS documentation, and I had genuinely never heard of it before…

Representing fold (sequential accumulation) as map composed with a reduce over a monoid, non-trivially and efficiently. This is a bedrock of parallel programming: however, although the general principle is known, applying it to concrete problems requires ingenuity

Image by Mohammad Bagher Adib Behrooz on Unsplash Why GraphQL for data engineers, you might ask? GraphQL solved the problem of providing a distinct interface for each client by unifying it to a single API for all clients such as web, mobile, web apps. The same challenge we’re now facing in the data world, where we integrate multiple clients with numerous backend systems. So what is GraphQL? In the world of microservices and web apps, GraphQL is a popular query language and serves as a data layer. It is essentially SQL on steroids for APIs. In this article, we go through how to combine the data from all services into a single, unified API.

The struggle of having a single performant, secure, and reliable data endpoint is real.

Container registries look simple until you need to debug what was actually pushed, why a pull picked the wrong image, or why deleting a tag didn't remove anything. Learn how registries work by pushing, pulling, inspecting, and deleting image data directly through the Registry API.

Seven years of maintaining a popular open source metasearch engine taught me a great deal about privacy, community, and the fundamental limits of the metasearch approach. Here is what led me to build Hister.

I’ve always found serverless platforms a little magical. You deploy a function, send a request, and somewhere in the world, resources are…

A small tool that captures Discourse plugin UIs in CI and publishes a gallery, modeled on Penar's theme-screenshots project but pointed at plugin routes.

I turned Telegram into a homelab control center, and it's made managing my Docker containers so much easier.

Tracking Chaos: Building a Real-Time Flight Anomaly Engine with Django, Celery, and Machine Learning Imagine walking outside on a quiet afternoon. You hear a sharp roar overhead, pull out your phone …

The background behind a new test suite for a compiler and a VM where test scenarios are all written in Markdown for both human and AI consumption

Old technology is any tech that’s, well… old. Small technology is any tech that has a small footprint: doesn’t require a powerful machine to run, doesn’t have a lot of bloat, doesn’t have anti-features like spyware and tracking. Technology that is old enough is almost always small because, by modern standards, that’s all that was possible back then. Some small tech is old, some is modern. Embedded systems are an example of modern small tech development in many cases.

UUCP is a system for exchanging data and requesting remote execution. It dates back to 1979, and was primarily used over Modems using telephone landlines for most of its days of popularity. It is an Asynchronous Communication system, which transmits data from one machine to the next on the way to its destination. Each intermediate node may store the data before passing it on to the next. Before dedicated Internet lines were widely available, UUCP was used to send Usenet and Email messages.