Explores inference efficiency in AI, focusing on preserving attention by reducing unnecessary reasoning steps and improving tooling design.
Explores how hypermedia and HATEOAS in REST APIs may become crucial for AI agents, with insights from industry experts on fixing broken web APIs.
Wolverine.HTTP adds support for the HTTP QUERY method via a new [WolverineQuery] attribute, enabling safe, idempotent search endpoints with request bodies.
A defense of polyfills in web development, arguing they are a net positive despite concerns from standards editors about design flexibility.
An interview discussing API specifications, OpenAPI vs API Blueprint, and how specs capture design intent.
An analysis of AI personas created by Kenneth Reitz, exploring the Digital Ouija Effect and emergent properties of language models.
Explains why API versioning should be a last resort, focusing on change management and compatibility rules to evolve APIs without breaking clients.
A developer argues that writing code by hand first, then using AI for maintenance, is more effective than AI generating initial drafts.
Explores idempotency in system design, its patterns, and critical role in reliable distributed systems, generative AI, and AI agents.
How to standardize resource naming in TypeScript monorepos using Google's API Design Guide convention.
PyTheory is a Python library for music theory, enabling chord identification from fret positions on any stringed instrument.
PyTheory is a Python music theory library built with a 'for humans' philosophy, overcoming a five-year creative block through human-AI collaboration.
Kenneth Reitz ports his personal site from Flask to his own web framework Responder, discussing the process and philosophy behind dogfooding.
Article discusses the true purpose of code review beyond bug-finding, emphasizing judgment, communication, and codebase health.
Explores when to choose stateless vs. stateful system design, discussing trade-offs for scalability, complexity, and use cases like shopping carts.
Argues that functions which modify data, like Python's random.shuffle(), should return new values instead of changing them in-place.
A response to critiques of the JavaScript Streams Standard, discussing its design philosophy, performance, and role as a web platform primitive.
Explains how to safely evolve data schemas using API-like discipline to prevent breaking downstream systems like dashboards and ML pipelines.
Explains why MCP servers often fail and provides best practices for building effective MCP servers by treating them as AI agent interfaces, not REST API wrappers.
Explains the concept of statelessness in REST APIs, contrasting it with stateful systems and detailing its benefits for scalability.
