Serge’s Substack

Serge’s Substack

Home
Notes
Archive
About
A year ago, Kimi's founder described this week
I watched a viral clip and hit a year-old interview that maps onto what shipped this month. The snow mountain, the two stones, and why one model was…
 • Serge Bulaev
I Had 11 AI Subagents Test Every PPTX Skill for Claude Code. Here's What Actually Works.
The same 5-slide pitch deck brief, eleven presentation skills, eleven parallel Claude subagents, zero mercy. Screenshots included.
 • Serge Bulaev
Guardrails for a Company of Cyborgs
An AI company from NASDAQ asked us to let their employees post with AI without burning the brand or the ticker. Here is the system we designed, and why…
 • Serge Bulaev
The New Cyborgs Don't Have Implants
We waited 65 years for the chrome arm. The merge happened in our messengers instead. I noticed it in my own business, by accident.
 • Serge Bulaev
I spent a month turning LinkedIn into software
Not a post generator. A ten-skill system for the whole grind, from picking an angle to figuring out who actually engaged. Open-source, MIT, free.
 • Serge Bulaev

November 2025

Why Model Distillation Matters - Part 2: When It Became Geopolitical
In Part 1, we covered what model distillation is, how OpenAI’s tools work, and what teams learned from a year of implementation.
 • Serge Bulaev
Why Model Distillation Matters (And How It’s Playing Out One Year Later) - Part 1
A year ago, at OpenAI DevDay 2024, they announced tools for model distillation that seemed like they’d change how people build AI products.
 • Serge Bulaev
Simon Eskildsen: Building a Learning Machine in the Age of AI
I recently watched another podcast about what I love most: memory, learning, AI, and how these things are reshaping our brains.
 • Serge Bulaev
When Three AIs Spent a Week Building Minecraft Civilizations (And One Jumped Off a Cliff)
What happens when you give three of the world’s most advanced AI models virtual bodies and set them loose in Minecraft?
 • Serge Bulaev
Can AI Predict Human Behavior? Stanford Proved It Can
Researchers at Stanford ran one of the biggest tests ever of whether AI can predict how real people will behave.
 • Serge Bulaev

October 2025

© 2026 Serge Bulaev · Privacy ∙ Terms ∙ Collection notice
Start your SubstackGet the app
Substack is the home for great culture