About
Hi, I’m Alexander Percivall, an enterprise architect at Google Cloud based in Stockholm, Sweden. I spend my days helping partners and customers build better solutions with cloud-native technologies, and my nights building AI agents that have (honestly) taken over my life.
The Short Version
I’ve been in tech for over 15 years. Started in consulting as a software engineer, worked my way up to lead developer and architect, then spent 5 years at Microsoft before landing at Google Cloud, where I’ve been for the past 8 years.
I’m extremely hands-on. If I’m recommending something to a customer, I’ve probably already built it myself. Sometimes at 2am, sometimes with more coffee than is medically advisable.
What I’m Building
Right now, I’m obsessed with AI agents. Not just chatbots that answer questions, but systems that actually do things proactively.
Pip: My attempt at building a real personal assistant. Not “ask me anything” but “I’ll keep things on track for you.” Integrated with email, calendar, files, messages. The assistant I wish existed.
Agent Orchestration Platform: Making autonomous agents observable and predictable for SWE and SRE work. Because “it mostly works” isn’t good enough when agents are writing your code.
I also run my own Kubernetes cluster because apparently I enjoy making my own life harder. And I tinker with my diabetes medical equipment, exploring how AI can help with my own treatment (either innovative or reckless, depending on who you ask).
Why This Blog
I work at the company that created the Transformer (that’s the T in GPT). I’m close to cutting-edge research and spend 10+ hours a day across AI use cases, both at work and on personal projects. I see a lot of things. Some of them work. Many of them don’t.
I want to share those learnings. The failures especially. That’s where the real lessons hide.
One of my favorite talks I’ve given was about the intersection of AI, humans, and process. I shared two failed projects and one success. The punchline? Tech wasn’t the issue. It rarely is.
That’s the perspective I bring here. Yes, the technology matters. But the people and the process matter more.
Who This Is For
Anyone interested in AI and agentic systems: how they work, where they fail, and how to actually implement them.
You don’t need to be a developer. Some of the most interesting conversations I have are with people who use AI in their everyday lives, not just those who build it.
I’m also here to learn from you. This isn’t a lecture; it’s a conversation. If you’ve got a story, I want to hear it.
Beyond the Keyboard
When I’m not arguing with language models, you’ll find me:
- Watching movies: I quote films constantly and collect physical media like it’s still 2005. Some of the best stories ever told are on screen (and in video games, if we’re being honest).
- Playing tennis: Badly, but with enthusiasm.
- Learning golf: New hobby. The learning curve is humbling.
- Walking Chewie: He’s a big black poodle and a lovable fool.
I also have two teenagers, which means I get a front-row seat to how the next generation actually uses technology. It’s equal parts terrifying and inspiring.
Let’s Connect
If any of this resonates, or if you think I’m completely wrong about something, I’d love to hear from you.
Check out the Contact page, or find me on LinkedIn and GitHub.