Boon Kgim

Hi, I'm Boon Kgim.

Cut through the AI noise. Practical AI insights from a builder who actually ships software for 19 years. AI agents, vibe coding, growth, marketing.
Founder of nanogent.ai, learnparrot.ai. Serial Technopreneur.


Recent Posts

All posts
The web and desktop apps of Claude and Codex are getting better.

The web and desktop apps of Claude and Codex are getting better.

Three reasons.

Vibe Coding
Most AI readiness tests skip the one layer that decides whether AI sticks: your people.

Most AI readiness tests skip the one layer that decides whether AI sticks: your people.

They go straight to data and process readiness. Important, yes. But I have watched plenty of transformations stall long before data or process ever became th...

Vibe Coding
Last Friday, I ran the Foundations of Claude workshop for Sequoia Group, a leadership and OD...

Last Friday, I ran the Foundations of Claude workshop for Sequoia Group, a leadership and OD...

As Sequoians, they believe in going high tech so that they can go high touch. They have been using AI, but they would like to learn how to use it better.

AI Agents
Why Is Innovation Harder in a Traditional Organisation Than a Startup? And What Business Leaders Can Do About It

Why Is Innovation Harder in a Traditional Organisation Than a Startup? And What Business Leaders Can Do About It

Innovation does not stall in traditional organisations because the people are less capable. It stalls because the risk and reward do not add up for the indiv...

AI Strategy
Ten more websites went live this week from my Foundations of Claude Code workshop.

Ten more websites went live this week from my Foundations of Claude Code workshop.

Check out what they built and vote for your favourite in the comments to help them win a portable monitor.

Vibe Coding
I am disappointed that Opus 4.8 still breaks one of the most basic rules in database design.

I am disappointed that Opus 4.8 still breaks one of the most basic rules in database design.

8 months ago, I noticed Opus 4.5 was not good at database design. So I built a DB design skill loaded with best practices.

Vibe Coding
AI trap: treating polished as correct.

AI trap: treating polished as correct.

Before AI, a polished report was a sign someone put in extra effort. Now it can be a sign they skipped it.

AI Agents
Claude.ai, Claude Desktop, Cowork, or Claude Code: which one should you use, and when? And why is Claude Code not only for coding?

Claude.ai, Claude Desktop, Cowork, or Claude Code: which one should you use, and when? And why is Claude Code not only for coding?

I get this question a lot, in workshops, in my DMs, and in offline conversations.

Vibe Coding

Practical insights from a builder who actually ships.

I have shipped nanogent.ai and built AI workflows that actually run in production. Here is what I learned across 100+ articles — filtered for practitioners, not theorists.

Everything I write comes from real implementation: the patterns that hold up, the ones that break, and the questions that no vendor demo will answer. Six pillars, one framework: build what works.

Cut through the AI noise.

Practical insights from a builder who actually ships.