Series

AI Models in Production: What Actually Matters

A 4-part series covering SWE-Bench rankings, open-weight models like MiniMax M2.5 and Qwen3.5, context window myths, and cost optimization with CRIT framework and model routing.

Open-Weight vs Proprietary: What SWE-Bench Verified Is Telling Us in 2026
I was looking at the SWE-Bench Verified leaderboard last week and the numbers surprised me. The gap between proprietary and open-weight models is almost gone. Not in some academic test. In actual bug
Apr 4, 20263 min read24
MiniMax M2.5 and Qwen3.5: The Open-Weight Models Worth Knowing About
In the last article I showed the SWE-Bench numbers. Open-weight models are basically tied with the proprietary ones now. Two models stood out to me: MiniMax M2.5 and Qwen3.5. Here's what I found out a
Apr 6, 20264 min read48
A Bigger Context Window Doesn't Mean Better Quality
When a company says their model has 1 million tokens of context, everyone assumes bigger is better. I thought the same thing. But after running LLMs in production for a while, I found out it's not tha
Apr 9, 20265 min read2
CRIT Framework and Model Routing: Spending Less on AI Without Losing Quality
In the previous articles I covered the SWE-Bench numbers, the open-weight models worth knowing, and why bigger context doesn't mean better output. Now I want to connect everything with two things I us
Apr 11, 20265 min read2

Command Palette