I'm an infrastructure engineer based in the Bay Area. Most of my time goes into Terraform, AWS, and the messy space where AI systems meet the infrastructure running them.
This is where I write down what I find. Technical for now — more of me will show up here eventually.
What Changes When a Model Version Ships
When a provider ships a new model version, your prompts run against a different model. We ran the same nine prompts across four model versions and measured what changed.
The Blind Spot in LLM Observability
Every observability tool promises visibility into your dependencies. I wanted to find out where that promise breaks down for LLM calls.
Inference Layer Cost: Routing by Cost Alone Isn't Enough
Three articles and four experiments later, I have more data than I started with. What I didn't expect is that every time more data was collected, it would make the question harder to answer, not easier.
Inference Layer Architecture: What the Experiments Actually Showed
Using LLMs for my own work, the same question kept surfacing: not whether the API responded, but whether what came back was actually right.
When Does Your Inference Layer Become a DevOps Problem?
I ran 1,000 identical API calls across five time windows and timed a provider migration twice: with an abstraction layer and without. The latency finding was expected. The behaviour change wasn't.
Evaluating Claude's Terraform skills: token cost, output quality, and when to use none of them
I tested four configurations (three MCP skills plus bare Claude) against five real infrastructure tasks. The token math is real, but it's not the most interesting finding.