About this blog

I'm Alexander Percivall, and I build AI agents every day. Some of them even work.

This blog is where I share what I'm learning - the patterns that actually work, the failures, and the surprisingly human side of working with AI agents.

What I Do

I live at the intersection of AI research and real-world implementation:

Building agentic systems with enterprise teams
Speaking about AI and agent design
Researching how humans and agents can actually work together
Making mistakes in production so you don't have to

Why This Blog Exists

Here's the thing: everyone building AI agents makes the same mistakes. We treat them like regular software (they're not). We forget about the humans using them (bad idea). We miss the patterns that only show up when real people start using our systems.

I've made every one of these mistakes. I've watched brilliant teams make them too. This blog exists to save you some pain.

What You'll Find Here

Real stories from the field - successes, failures, and the unexpected stuff in between.

Practical patterns for when humans and agents need to collaborate (which is always).

Complex ideas made simple - because if I can't explain it clearly, I don't understand it well enough.

Honest takes on what's working, what's hype, and what we're all still figuring out.

Who This Is For

You don't need to be technical to read this. Whether you're:

Building AI agents or thinking about it
Leading a team navigating AI
Curious about how humans and machines actually work together
Trying to separate signal from noise

If you're wondering how AI agents work in the real world (not just in demos), this is for you.

I'm a Partner Engineer at Google. Everything here is my own thinking, my own words, and my own mistakes.