Practical AI

DeepSeek’s Vision Model Exposes the Dirty Secret of AI: AGI Is a Distraction

DeepSeek’s founder said they’d go text-only to achieve AGI. Then they shipped a vision model. This isn’t hypocrisyβ€”it’s the industry’s dirty secret: even the most principled AI labs are forced by market demand to build practical multimodal tools. The real race isn’t about abstract intelligence; it’s about reading screenshots reliably.

I Spent 30 Days Ignoring AI Hype. Here’s What I Learned About Loop vs. Graph Engineering.

The AI industry is a hype machine that recycles old concepts under new names. Agent Loop and Graph Engineering are not competitors. The Loop is an organizational solution for single-agent self-correction. The Graph is an architectural solution for multi-agent data flow. Adopting the Graph prematurely reveals hidden complexity and burns tokens. Learn slowly, and you realize you don’t have to learn anything new.

The AI Model That Won the Only Race That Matters: Not Being Annoying

A head-to-head test between GLM 5.2 and GPT-5.6 Sol reveals a surprising winner. It wasn’t about raw intelligence β€” it was about which model caused less frustration. GPT-5.6 Sol dominated on instruction-following and formatting, proving that the new AI moat isn’t capability, but non-annoyance. The model that wins your workflow is the one that doesn’t make you correct its mistakes.