01
1996
A first byline and a first shared web project
At 14, I published an article in Centennial Middle School’s online newspaper and helped build ThinkQuest astronomy entry 3516 with two other students.
Product · infrastructure · working systems
I build hard systems, then make them understandable.
I started building on the early web at fourteen. Since then I have worked through enterprise software, Azure Blob, the storage path behind demanding AI workloads, and the agent systems I run today. The technology changed. My way of learning it did not.
How the work changed
The same method, four systems
- 01Early web
Student publishing and shared web projects.
- 02Independent builder
Web products, automation, and open technical writing.
- 03Cloud infrastructure
Azure Storage product leadership and frontier-scale customer work.
- 04Applied AI
VAST, Verdify, and bounded agent systems.
30 yearssince my first published byline
12½ yearsat Microsoft, ending in Azure Blob
Named by OpenAIunder Microsoft Azure Support
Longmont rootedBoulder County native, resident since nine
What stayed the same
I learn by building.
At fourteen, that meant a student newspaper and a shared astronomy site. At sixteen, it meant Perl and Linux. Later it meant a web company, SharePoint, Azure Storage, and the infrastructure underneath AI. Today I operate a fleet of coding agents and build a greenhouse that lets AI propose changes without giving it direct control of the hardware.
A working life
Four points on a much longer timeline
02
2003
Longmont City Council candidate
At 21, I ran in Ward III and received a reconciled 1,331 votes, or 28.24 percent, against an incumbent.
03
2017–2025
Azure Storage product leadership
I led strategic customer engagement and Blob experience product work, finishing with the public-facing title Group Product Manager, Azure Blob Storage.
04
Oct 2025–present
VP, Cloud Product Management
I connect VAST’s data platform to cloud product strategy and the way hyperscalers, neoclouds, and AI workloads actually operate.
The OpenAI chapter
My work was under the models.
From 2017 through 2025, I led Azure Blob product work, including the Azure-side storage relationship supporting OpenAI and Microsoft AI. OpenAI’s 2019 Dota 2 report names me under Microsoft Azure Support. At SNIA SDC 2024, Jegan Devaraju and I explained how Azure Blob supported large-scale model training.
My part was product and infrastructure work, alongside a much larger team.
Read the Azure and OpenAI chapter“I came home from my first meeting with OpenAI in 2017… mind-blown.”
Selected writing
I write to understand how things work.
The collection currently includes 39 articles: 32 technical pieces and 7 essays. It begins with writing from earlier versions of vallery.net and stays open for whatever I publish next.
Jan 20, 2008
Extracting emails from Gmail and Google Apps for Domains
I wanted the useful contact history buried in my Gmail mailbox, so I built a tool to extract, rank, deduplicate, and export email addresses. That utility later became MineMyMail.
Feb 1, 2008
Automatically backup your Mac to Amazon S3
Time Machine protected me from a failed disk, but not a failed house. I combined local backups with Amazon S3 to create an automated off-site copy.
Mar 27, 2007
What is Web 3.0? A review of the ICWSM
I used the 2007 International Conference on Weblogs and Social Media to ask what people meant by Web 3.0. My answer focused on social data, identity, and machines learning from human behavior.
What I am building now
Cloud data, coding agents with guardrails, and a greenhouse that has to survive the weather.
I lead cloud product work at VAST Data, build Verdify with my adult son, and operate a private agent platform on Kubernetes. All three need clear limits, observable behavior, and a person who remains accountable for the result.