Max HenryFounder & Executive Director, Global Supply Chain Council (GSCC) · Founder & Executive Director at Global Supply Chain Council (GSCC)

AI Coding Tools: My CloudCode & Opus 4.5 Experience #shorts

26 Jun 2026 · 2 min

Watch on YouTube

2:02

In this short, a product manager recounts a focused experiment with AI‑assisted coding tools undertaken at the end of the previous year. Feeling a professional need to stay current, they first tried Lovable, noting its ability to spin up a complete application—including user management and a database—rapidly. Seeking more technical flexibility, especially for a graph‑database‑based supply‑chain prototype, they moved to Claude Code. The timing coincided with the release of Opus 4.5, which the speaker describes as an inflection point that dramatically improved Claude Code’s performance. The experience convinced them that powerful, agentic AI will extend beyond programming into broader knowledge‑work domains such as network design, optimization, and supply‑chain analytics. They are now actively building these ideas using CloudCode.

Chapters

In this job, end of last year, I started to experiment heavily with AI vibe coding tools. Really took some time off the job actually to experiment there because I felt like, or I was afraid to be left behind and felt a professional necessity to know what's going on, especially as a product manager. You really know, need to know what's going on in this field. So I started with Lovable. I was quite amazed by how quickly one can set up a whole application including user management and a database and everything. But I wanted to try out other ideas, more technical ideas like using a graph database for supply chain at that time. That was outside the limitations of Lovable. That's why I tried Claude Code. I've been super lucky, I think, with the timing when I tried Claude Code for the first time because this was a few days after the release of Opus 4.5, which was the inflection point for Claude Code, I believe. At that time, it got really, really good. I think the talk about it also started. I realized quickly how amazing it works. I mean, I've been programming myself a bit and it really changed a lot for me. And I also started to think about the consequences of such powerful tooling and became convinced that it will not stop at programming. Like this agentic AI can have many different use cases, will be used in all parts of knowledge work as also network design, network optimization and also supply chain analytics in general. So I really began to have a vision how this could look like and I started to bring it to life using mostly cloud code.