Thoughts on infrastructure, automation, AI systems, and building technology that scales. Practical insights from real-world implementations.
From studying Kurzweil at UCLA to building AI infrastructure—reflections on technological acceleration, American leadership, and why compute capability matters for the future we're creating.
Design patterns for scaling AI compute, H100/H200/B200 architecture, and token-per-watt optimization strategies for production inference workloads.
Designing automation frameworks that integrate disparate systems, reduce manual overhead, and scale across enterprises without creating technical debt.
Practical techniques for reducing latency and cost in production LLM deployments, from model quantization to intelligent request batching.
Lessons learned from managing complex infrastructure with Terraform and Ansible. Patterns for maintainability, testing, and team collaboration.
Get notified when new articles are published. No spam, just insights on infrastructure and automation.
Unsubscribe anytime. Your email is safe with me.