BYOB: Build Your Own Benchmark

AI benchmarks are saturating, getting harder to verify, and increasingly irrelevant to how most people use models. The replacements are weirder - and…

Mar 1 • Charlie Guo

22

6

3

February 2026

The Emerging "Harness Engineering" Playbook

The converging best practices for building with coding agents, from OpenAI to Stripe to OpenClaw.

Feb 22 • Charlie Guo

68

11

7

GPT-5.3-Codex and Claude Opus 4.6: More System Card Shenanigans

Evaluation awareness, reward hacking, and the cybersecurity problem.

Feb 11 • Charlie Guo

33

5

3

The Codex App Has Upended My Daily Workflow

I haven't opened my IDE in days. I'm not sure I miss it.

Feb 2 • Charlie Guo

28

7

2

January 2026

Humans Welcome to Observe

The OpenClaw/Clawdbot explainer: personal AI agents, security nightmares, and robot religion.

Jan 31 • Charlie Guo

44

5

8

Skills, Tools and MCPs - What’s The Difference?

The primitives for AI systems are still being invented. Here's where we are today.

Jan 24 • Charlie Guo

44

4

5

The AI Manager's Schedule

The new skills you need when your reports are LLMs.

Jan 16 • Charlie Guo

27

1

3

On Joining OpenAI

And the next chapter of Artificial Ignorance.

Jan 9 • Charlie Guo

34

36

1

10 AI Stories That Shaped 2025

A look back at the year in AI news.

Jan 4 • Charlie Guo

20

2

2

December 2025

AI Roundup 150: Empire State Defiance

December 27, 2025.

Dec 27, 2025 • Charlie Guo

21

2

2

AI Roundup 149: Flash Forward

December 20, 2025.

Dec 20, 2025 • Charlie Guo

18

2

1

AI Trends for 2026

And a report card for last year’s predictions.

Dec 17, 2025 • Charlie Guo

34

5

7