Community·
The Month I Almost Quit
Every founder has a breaking point. Mine came in September—returning from vacation to an AI landscape that had exploded without me. Here's what happened next.
Insights & Updates
Blog
14 posts
Community·
Marketing a Developer Tool When You're Also the Developer
The old paradox was: every hour marketing is an hour not coding. AI broke that trade-off. Here's how I market while AI builds.
Community·
Planning vs. Building: Finding the 80/20 for Solo Founders
Writing specs is the new superpower. But how much planning is too much? Here's how I actually split my time—and what I learned about productive procrastination.
Community·
When Not to Use Kodebase
Kodebase isn't right for every project. Here's an honest assessment of when executable documentation adds value—and when it's overkill.
Community·
I Built Kodebase in a Month with AI: What I'd Do Differently
The metrics look impressive. But behind them are mistakes, wrong turns, and lessons learned the hard way. Here's my honest retrospective.
Engineering·
Git is the New Database: Why We Built a Project Intelligence System that Eliminates Context Decay
By treating Git as the single source of truth for development intelligence, Kodebase gives every decision, artifact, and specification the superpowers of version control: immutable history, offline access, and zero infrastructure cost.
Engineering·
How to Write Specs AI Can Actually Execute
AI fails not because it's dumb, but because your specs are vague. Learn how to describe what you want in plain language—and let the system handle the technical translation.
Engineering·
The Scout, Sherpa, and Monk Trinity: A Founder's Guide to AI-Led Development
An enforced system of three specialized MCP tools—Scout, Sherpa, and Monk—designed to capture, execute, and learn from a founder's vision, transforming domain expertise into production-quality products.
Engineering·
Memory Isn't Context: Why Universal LLM Memory Systems Fail
New benchmarks show 'smart' memory systems like Mem0 and Zep perform worse than simply feeding context directly. Kodebase takes a different approach entirely.
Engineering·
The Coming Split: AI-Native Startups vs. Everyone Else
A new class of startups is emerging—built from day one with AI at the core. They operate differently. They scale differently. And they're about to make traditional startups look slow.