Prometheus Cardinality: The Label Decisions That Decide Whether Your Monitoring Survives
Every Prometheus failure I have debugged that was not a network problem eventually came down to the same thing: too
From Our Blog
Constrained Decoding: How Structured LLM Output Actually Works
Ask an LLM for JSON and you will usually get JSON. “Usually” is the word that ruins your week. One
Continue readingConstrained Decoding: How Structured LLM Output Actually Works
HTTPX2 and the Art of the Open-Source Handover
In late August, OpenAI’s Python SDK — one of the most-installed packages on PyPI — quietly completed a migration to
Continue readingHTTPX2 and the Art of the Open-Source Handover
5 Trending GitHub Repos: Virtual iPhones, Local Stem Separation, and a Rust KV Store That Documents Its Crash Guarantees
The last week of August delivered an unusually varied batch of open-source launches. A tool that boots a virtual iPhone
Native Sidecar Containers in Kubernetes: Ordering, Jobs, and the End of Shutdown Races
The sidecar pattern has been part of the Kubernetes toolkit for as long as pods have existed: run a second
View Transitions in Practice: Native Page Transitions for Multi-Page Sites
For most of the last decade, smooth page transitions were a privilege reserved for single-page applications. If you shipped a
Continue readingView Transitions in Practice: Native Page Transitions for Multi-Page Sites
Idempotency Keys: Making API Retries Safe by Design
The network drops the connection two seconds after your client sends the payment request. Did the charge go through? Your
Continue readingIdempotency Keys: Making API Retries Safe by Design
Dead Letter Queues in Kafka: Stop Poison Pills Before They Stop You
A message arrives that your consumer cannot parse. Maybe the producer shipped a new schema without telling anyone, maybe the
Continue readingDead Letter Queues in Kafka: Stop Poison Pills Before They Stop You
The keyv Worm and the Anatomy of a Modern Supply Chain Attack
On August 4, 2026, someone took over the GitHub account of the maintainer of keyv, a key-value storage abstraction library
Continue readingThe keyv Worm and the Anatomy of a Modern Supply Chain Attack
Speculative Decoding in vLLM: Draft, Verify, and Cut Latency Without Losing the Distribution
Large language model inference has an awkward performance profile: the GPU does enormous math, then waits. Every token requires a