Blog
Latest articles
4 min readClaude Code Sends 33k Tokens Before Reading the Prompt; OpenCode Sends 7k: A Practical Overview
Claude Code sends 33k tokens before reading your prompt; OpenCode sends 7k. This overhead affects cost and latency in agentic coding workflows. We break down the technical reasons and practical implications.
6 min readI Used Claude Code to Get a Second Opinion on My MRI: A Practical Overview
A developer fed his MRI scan to Claude Code and got a second opinion. Here's how the experiment worked, what it cost, and why you shouldn't rely on it for medical decisions.
4 min readWhat is Claude Code is steganographically marking requests? A Practical Overview
Claude Code steganographically marks requests by embedding invisible patterns in prompts to trace misuse. Here's how it works, the technical tradeoffs, and what developers should know.
4 min readBuilding reliable agentic AI systems: A Practical Overview
A practical guide to building reliable agentic AI systems covering structured outputs, observability, fallbacks, and cost controls with real code examples.
4 min readTokenomics: Quantifying Where Tokens Are Used in Agentic Software Engineering
Tokenomics quantifies token usage per step in agentic software engineering. This post breaks down the numbers, tradeoffs, and practical tips for cost optimization.
4 min readWhat is SpaceX Is Buying Cursor? A Practical Overview
SpaceX is buying Cursor, the AI-powered code editor. The deal signals a shift in how AI coding tools are valued and deployed. Here's a practical breakdown of what's happening and what it means for developers.
4 min readOpen Code Review – An AI-powered code review CLI tool: A Practical Overview
Open Code Review is an open-source CLI tool from Alibaba that uses AI to review code changes. It runs locally, supports multiple LLMs, and costs about $0.01 per review. Here's a practical breakdown.
5 min readWhat is Harness engineering: Leveraging Codex in an agent-first world? A Practical Overview
Harness engineering is the practice of building structured, safe environments for AI agents to execute code. This post explains how to leverage OpenAI Codex in an agent-first world, with concrete examples, cost breakdowns, and honest tradeoffs.
3 min readWhat Is the Text in Claude Code's Extended Thinking Output? A Practical Overview
Claude Code's Extended Thinking output reveals the model's internal reasoning chain. But as Patrick McCanna points out, this text is not a faithful transcript of the model's thought process. Here's what developers need to know.
3 min readHow to Build with Codex: Avoiding the Logging Bug That Writes TBs to Local SSDs
A logging bug in Codex can write terabytes of logs to your local SSD. Here's how to detect, fix, and prevent it with practical steps and configuration changes.
4 min readWhat Is Failing Grades Soar with AI Usage, Dwindling Math Skills in Berkeley CS Classes? A Practical Overview
UC Berkeley CS professors report a sharp increase in failing grades alongside higher AI usage and declining math skills. This post breaks down the data, the likely causes, and what it means for hiring and education.
4 min readWhat is My Agent Skill for Test-Driven Development? A Practical Overview
An AI agent skill for test-driven development automates the red-green-refactor cycle. Here's how it works, what it costs, and when to use it.