Writing

Essays on applied AI — agents, evaluation, retrieval, developer tooling — plus engineering craft. 19 posts. Shorter thoughts live in notes; new here? Start with the curated path.

Featured

Confusion matrix of LLM-judge verdict vs. real outcome, with the false-pass cell highlighted
Evaluation

Ground-Truthing Your LLM Judge: Does Your Eval Actually Track Reality?

LLM-as-judge is the default eval now — but a judge is just another unreliable model. Here is how to check whether its verdicts correlate with real outcomes, and catch the leniency bias most judges have.
Photo by park ingyeom on Unsplash
Agents

I Built 50 AI Skills for the Parts of My Day I Never Got Around To

The biggest productivity gap isn't code — it's context gathering, note-taking, and status updates. I built 50 AI workflow skills to make those 15-minute tasks actually happen every day.
Weights
Evaluation

Reward Engineering and Evaluation: Making Your System Learn What Matters

This is the third post in our series on probabilistic prompt pipelines. In our first post, we explored why static prompts become bottlenecks. The second…
cart-unsplash
Retrieval

Building the Selector: Retrieval, Bandits, and Cold Start Solutions

This is the second post in our series on probabilistic prompt pipelines. In the first post, we explored why static prompts become bottlenecks and saw a…
blue sign
Retrieval

Why Static Prompts Fail and How Probabilistic Selection Solves Real Problems

This is the first post in a four-part series on building production-ready probabilistic prompt pipelines. By the end of this post, you'll understand why…
apple II
Agents

Claude Code: The Hidden Testing Ground for AI Agents

I've been using Claude Code daily since it launched for general use. After months of working with it, I've developed a theory about what's actually…
Horses and Grooms Crossing a River
Agents

Staying Sane on a 6,000-Line Refactor with Claude Code

As large language models become more capable, it's tempting to rely on them for increasingly complex engineering tasks. Tools like Anthropic's Claude Code…
prompting power
Retrieval

Enhancing Retrieval-Augmented Generation with XML Prompting

In my current role, I’ve been working on a Retrieval-Augmented Generation (RAG) system designed to provide safe, reliable personal finance guidance. Given…

More writing

Developer Tools

Use /insights to Improve Your Claude Code Setup

I ran /insights in Claude Code and walked through what it found — friction patterns I hadn't noticed, CLAUDE.md rules I should have written weeks ago, and what changed after I applied the suggestions.
Career

One Thing: Balancing Hard and Soft Skills in the Age of AI

In “Formal Methods Aren’t Just Formalities”, Philip Brooker dives into the importance of applying formal methods to complex software systems. These…
Developer Tools

One Thing: Cursor, Claude, and the Mocking Struggle

As someone who spends a lot of time coding, I rely heavily on Cursor as my daily IDE, and I most often use Claude 3.5 Sonnet as my preferred model for…
Automation

Automating Meeting Note Management with Krisp, Zapier, Google Drive, and Notion

Transform how you manage meeting notes with an automated workflow using Krisp, Zapier, Google Drive, and Notion. This step-by-step guide will help you capture, store, and organize your meeting transcripts effortlessly, saving time and ensuring all your notes are neatly summarized and accessible. Ready to simplify your workflow? Let’s dive in!

Engineering fundamentals

TypeScript

Mastering Conditional Types in TypeScript

As your TypeScript codebase grows, so does the need for abstractions that are both expressive and type-safe. Conditional types unlock a form of type-level…
React

Understanding useEffect vs. useLayoutEffect in React: A Comprehensive Guide

React Hooks have profoundly changed the way we write React applications, allowing developers to manage state and side effects elegantly in functional…
React

Why Returning Cleanup Functions in useEffect Matters in React

React’s useEffect hook is one of the most powerful and commonly used features in functional components. It enables side effects like data fetching,…
Infrastructure

A Guide to Setting Up Vercel Log Storage in AWS S3

Modern cloud applications generate vast amounts of logs that are crucial for debugging, monitoring, and maintaining application health. While Vercel…

From the archive

Personal

One Thing: Kneading a New Hobby

I’m kicking off a new blogging experiment I’m calling “One Thing.” Each day, I plan to share one thing I learned—big or small, tech-related or totally…
Developer Tools

Creating A Blog With GatsbyJS - Part Two

Create a blog with GatsbyJS
Developer Tools

Creating A Blog With GatsbyJS - Part One

Create a blog with GatsbyJS