Blog
Notes on AI inference, cloud cost and platform engineering for small teams.
-
> API gateways: what they do, where they go, and what they cost to run
“Do we need an API gateway?” It depends less on the product than on which problems you actually have, and the cost of getting it wrong ranges from a few pounds a month to a few thousand.
-
> Learning a Codebase with LLMs - What Each Model Tier Buys You (October '26)
I have a toolset designed to help me understand codebases and application design. I wanted to understand which model was best suited to the job, and which gave the best value for money. I tested two dozen language models. The cheapest decent model got me 80% of the way for half a cent. The last stretch didn’t come from a bigger model. It came from the harness around it.
-
> What an inference gateway is, and why a small team needs one
If your application calls a model provider’s API directly, the provider is baked into your code. Every model change is a code change. Every price rise lands straight on your bill. You find out about an outage when your users do.