Infrastructure engineer - UK. Twenty-odd years of building the plumbing.
I'm building Fusewire, a self-hosted governance proxy that hard-stops LLM API overspend before a request ever leaves your network. It runs in your VPC, never holds your keys and never touches the payload. Pre-launch, in internal testing.
Alongside the build, I'm running a research project: how do teams operating LLMs in production (especially in regulated environments) actually handle runaway spend and "prompt data can't leave our environment" constraints?
The format is a 20-minute conversation about your setup and war stories. It is not a sales call - there's nothing to sell yet. I aggregate what I learn and share the patterns back with everyone who takes part.
If your AI bill has ever surprised you, I'd like to hear about it: thedanielanthony@gmail.com
I'm a 20-year software veteran and startup CTO who spent six years owning cloud spend and production infrastructure for live AI and gaming platforms - including a Kubernetes-based MMORPG backend where every millisecond of latency and every dollar of GCP bill was my problem. I've been building with LLMs since before "RAG" had a name and Fusewire exists because I needed it: real-time budget enforcement that never sees your prompts, never holds your keys and never rewrites a byte.