Laurent DeSegur

Software engineer building open-source AI agent infrastructure. Creator of KarnEvil9 — a deterministic AI agent runtime implementing Google DeepMind's Intelligent AI Delegation framework. Writing abou

New York, USA Joined on Mar 5, 2026 oldeucryptoboi@gmail.com https://oldeucryptoboi.com

Education

CS + EE

Work

Managing Director

Laurent DeSegur

Apr 8

How Tool Search Defers Tools to Save Tokens

#claudecode #mcp #architecture #aiagents

21 min read

Laurent DeSegur

Apr 8

How Claude Code Extends Itself: Skills, Hooks, Agents, and MCP

#claudecode #architecture #mcp #aiagents

18 min read

Laurent DeSegur

Apr 8

What Happens When Claude Code Calls the API

#api #claudecode #architecture #streaming

18 min read

Laurent DeSegur

Apr 7

94% Exposed, 30% Adopted: Why Engineering Leaders Need to Rethink How They Hire

#ai #hiring #softwareengineering #futureofwork

3 min read

Laurent DeSegur

Apr 7

How Claude Code Manages Infinite Conversations in a Finite Context Window

#ai #claudecode #architecture #devtools

20 min read

Laurent DeSegur

Apr 6

How Bash Command Safety Analysis Works in AI Systems

#security #ai #bash #devtools

5 min read

Laurent DeSegur

Apr 1

The Claude Code Leak: What Anthropic Accidentally Revealed About the Future of AI

#ai #claudecode #anthropic #agents

8 min read

Laurent DeSegur

Mar 31

OpenAI Just Shipped a Plugin So Codex Runs Inside Claude Code

#openai #codex #claudecode #devtools

3 min read

Laurent DeSegur

Mar 31

Inside Claude Code's Architecture: The Agentic Loop That Codes For You

#claude #ai #architecture #devtools

6 min read

DEV Community

Laurent DeSegur

Badges

Writing Debut

Currently learning

Currently hacking on

Available for

How Tool Search Defers Tools to Save Tokens

How Claude Code Extends Itself: Skills, Hooks, Agents, and MCP

What Happens When Claude Code Calls the API

94% Exposed, 30% Adopted: Why Engineering Leaders Need to Rethink How They Hire

How Claude Code Manages Infinite Conversations in a Finite Context Window

How Bash Command Safety Analysis Works in AI Systems

The Claude Code Leak: What Anthropic Accidentally Revealed About the Future of AI

OpenAI Just Shipped a Plugin So Codex Runs Inside Claude Code

Inside Claude Code's Architecture: The Agentic Loop That Codes For You