Running Llama-3 Behind Your Own Firewall: a vLLM Production Checklist
GPU memory headroom, continuous batching, API-key rotation, and the monitoring you need before an internal LLM endpoint counts as production.
Blog
Tech
Tech
Tech
Finance
Finance
Finance
Fitness
Fitness
Personal blog
Long-form writing on technology, finance, and fitness.
On the editorial calendar
Coming soon.
The API Proxy Gateway Pattern for LLM Traffic
Put one gateway between every app and every model — routing, budget caps, PII scrubbing, and audit logs in a single choke point.
Avoiding the Vulnerability De-listing Trap in Cloud Marketplaces
Why a marketplace image that passed certification last month can be de-listed today, and the CVE-freshening pipeline that prevents it.
SaaS Pricing Models for Infrastructure Software
Per-vCPU, per-node, flat-tier: what actually clears procurement in enterprise accounts, with real marketplace data.
Burning Down Committed Cloud Spend Without Wasting It
Committed-spend agreements expire whether you use them or not. A framework for routing real workloads at the burn-down.
Strategic Planning for a Family-Owned Technology Business
Annual planning that survives contact with reality: capital allocation, partner programs, and when to say no to growth.
Training Consistency When You Run a Company
The minimum-effective-dose program that survives travel weeks, launch weeks, and everything in between.
A Diet Plan You Can Actually Sustain at a Desk
Protein targets, meal timing around deep-work blocks, and what to keep out of the office entirely.