I write about AI, software engineering, and the systems that shape our world.
Sep 19, 2026· 13 min read
Does size matter? There. I asked it. Humanity has spent an unreasonable amount of time arguing about this question, so naturally we eventually gave the problem to artificial intelligence. Except now w
Sep 7, 2026· 10 min read
I would let GPT-6 Astra do more work. I would not let GPT-6 Astra decide what it is allowed to do. That distinction is more interesting to me than most of the benchmark numbers in OpenAI's launch. Op
Sep 5, 2026· 6 min read
The most visible AI product is a chat box. The most instructive AI systems often have no chat box at all. Look across Google DeepMind's work on algorithms, quantum error correction, robotics, and hist
Sep 1, 2026· 6 min read
An API migration can compile, return the expected text, and still be wrong. That is the real risk in moving from OpenAI's Assistants API to Responses. The old objects do not map to new objects without
Aug 30, 2026· 7 min read
Connecting an agent to a database does not remove hallucination risk. It gives the system a second source that can be stale, malformed, unauthorized, or simply wrong—and a protocol that may make the m
Aug 27, 2026· 6 min read
Agentic IDEs can produce a runnable full-stack application in minutes. That does not mean the application became cheap to trust. In a controlled study of 48 generated projects, the researchers spent a
Aug 25, 2026· 8 min read
A GitHub repository collecting supposedly “leaked” system prompts from ChatGPT, Claude, Gemini, Grok, and other AI products has crossed 63K+ stars and 10K+ forks. Most of the discussion naturally focu
Aug 24, 2026· 6 min read
“Which model is best for coding?” sounds like a precise engineering question. Usually, it is missing half of the system. A coding agent is a model inside a harness that chooses tools, assembles contex
Aug 22, 2026· 6 min read
An AI agent can pass the task and still fail the engineering evaluation. That happens when a context-compression policy preserves the final outcome but changes how the agent gets there. The agent stil
Aug 20, 2026· 9 min read
An AI agent can produce the correct benchmark answer and still fail the evaluation. That is what happens when the path to the answer crosses an unauthorized boundary. In July 2026, an autonomous agent
Aug 20, 2026· 22 min read
A coding agent can now inspect a repository, implement a feature, modify multiple files, run tests, fix failures, and prepare code for review. That is a remarkable change in software development. It a
Aug 8, 2026· 16 min read
I like Claude. That is precisely why I have a problem with Claude Pro. Claude can write exceptionally well. Its coding models are genuinely competitive. Claude Code is one of the strongest arguments
Aug 5, 2026· 3 min read
Intelligence isn't just about the model When we talk about AI products, we usually compare models. Qwen vs Gemini. Claude vs GPT-6. Reasoning vs Non-Reasoning. Latency vs Context window vs Benchma
Aug 5, 2026· 9 min read
A portfolio is often treated as a visual résumé: a polished landing page, a few project cards, and a contact link. I wanted mine to do more. This rebuild was about turning my portfolio into a durabl
Aug 4, 2026· 6 min read
A long context window sounds like an obvious advantage for an AI agent. The agent can retain more search results, tool outputs, intermediate reasoning, and evidence. Give it enough context, and perhap
Aug 4, 2026· 8 min read
Disclosure: This case study is anonymized. Product screenshots, source code, client branding, proprietary workflows, and internal metrics have been excluded. Industry: Health insurance Project type:
Aug 4, 2026· 10 min read
Disclosure: This case study is anonymized. Client branding, production screenshots, source code, internal architecture, customer data, and business metrics have been excluded. Industry: Travel insura
Aug 3, 2026· 12 min read
Disclosure: This case study is anonymized. The client’s identity, product branding, source code, production screenshots, internal architecture, customer information, commercial terms, and performance
Aug 3, 2026· 11 min read
Disclosure: This case study is anonymized. The client and employer relationship, production screenshots, source code, tenant information, employee records, internal architecture, and commercial metric
Aug 2, 2026· 5 min read
For most of my career, I thought software engineering was primarily about building things. Choosing the right architecture. Writing clean code. Designing scalable systems. Picking the right framew
Jul 28, 2026· 14 min read
Disclosure: This case study is anonymized. Product branding, source code, production screenshots, employee and candidate information, internal architecture, commercial details, and performance metrics
Jun 22, 2026· 5 min read
Every frontend developer has heard some variation of these. "It's just a screen." "Backend does the real work." "AI can generate that." "Just use ShadCN." "Just copy the component." "It's just a chec
Jun 18, 2026· 3 min read
More Thinking Doesn't Always Mean Better Answers Reasoning models are designed to think step by step, and the common assumption is simple: More thinking = better results. I expected reasoning models