GPT-5 Is the Best Coding Model in the World: Biggest Leap Since 3.5 Sonnet
GPT-5 doesn't just write code; it thinks like an engineer. When faced with complex dependency conflicts that stumped previous models, GPT-5 takes a methodical…
Read postAll LLM posts
Everything in this archive, newest first.
My LLMs Have Personalities and I Can’t Unsee It
Spending 8+ hours a day with AI assistants does things to you. Time to share my completely scientific personality assessments. GPT-3.5: The Chatterbox GPT-3.5…
Read postIn AI, $20/Month is the New Free Tier
Let's stop pretending $20/month gets you professional-grade AI access. That era is over. Running LLM inference is expensive, and the honeymoon phase is over.…
Read postWe’re not writing code anymore. We’re directing it.
The best mental model I've found for AI-assisted development: you're a film director. AI Coding is Directing. You have a vision in your head.…
Read postCoding with AI: The Limit is No Longer Intelligence, But Cost
For the first time, it's not the AI's capabilities that limit me. It's what I can afford to spend. I use AI for most…
Read postAnthropic’s Infrastructure Problem
Anthropic's Claude Sonnet 3.7 has established itself as one of the most capable AI models available today. But there's a critical problem that's been…
Read postShaping the Future, One Prompt at a Time
The saying "You are who you surround yourself with" takes on new meaning today. Many of us spend a lot of time talking to…
Read postFinding Long Files for Better Cursor IDE Performance
As a developer working with Cursor IDE, you've probably run into the line limit issue. Cursor has limitations on how large files can be…
Read postAlways Use Max Models in Cursor IDE
If you've been using Cursor IDE since its early days, you might have noticed a bizarre decline in quality around the time Claude 3.7…
Read postAI Soft Skills: The New Differentiator for Language Models
Raw intelligence is no longer the sole marker of a language model's utility. Cutting-edge models are reaching parity in problem-solving abilities. As a result,…
Read postSearch Will Rule the AI Era
ChatGPT's Deep Research is just the first phase. The real revolution will happen when the underlying research model becomes available through an API that…
Read postThe Best AI Model Doesn’t Run on NVIDIA Chips
Google's latest AI model Gemini 2.5 Pro has emerged as arguably the best-performing AI model to date. This experimental model leapfrogged the previous leader…
Read post