I'm terra tauri, enabling SRE at Bit Complete. I combine deep technical expertise with building high-performing teams. Previously at Grafana Labs supporting 10,000+ Kubernetes nodes. KubeCon EU speaker on eBPF observability.
The default answer to "I need more from my AI" is still a bigger model on a bigger GPU. But every piece of the other answer, horizontal scaling for inference, already exists in the research. Here's how small, specialized, cache-friendly workers make AI cheap instead of scarce.
Bolting AI onto an organization built for walking pace doesn't make it faster. The bottleneck was never the typing — it was coordination, and AI only shone a light on it.
Why coding agents need a sandbox the agent cannot reach, an audit log the agent cannot write, and signed instructions the agent cannot rewrite.
I enable SRE at Bit Complete, combining technical depth with building high-performing teams. Previously at Grafana Labs, I built network infrastructure supporting 10,000+ Kubernetes nodes, executed zero-downtime CNI migrations, and investigated softIRQ bottlenecks at 1M+ packets/second.
I'm also building AI-powered tooling (Claude Code Agent SDK), creating open-source projects (Nethax network testing framework), speaking at KubeCon, and experimenting with creative technical projects like this site.