I have been an engineer on Amazon Checkout since 2018. It is the purchase flow: hundreds of millions of people a day, over a million requests a second.
These days I lead technical work across several teams and build agents that take on software maintenance. Before that, I led Checkout engineering for Add to Delivery and a multi-year move from a monolith to Java microservices.
I live in Seattle. Outside software I read about ancient cities and go to stand in what is left of them.
Some things I’ve worked on
Agents for engineering work
2025–present
I designed and built a system that assigns maintenance tasks, such as removing expired feature flags, to agents. Agents are not deterministic: the same task can stall, produce a different change on a second run, or fail halfway through. So the system treats each task as a durable workflow rather than a single prompt. Every task is validated against an operating procedure before it starts, runs as a state machine that records each step, and can be resumed or retried from the last recorded step without repeating work already done. Outputs are evaluated before a human reviewer sees them, and nothing merges without that review.
I also built Model Context Protocol (MCP) tooling that lets agents investigate checkout failures across purchase-journey data and service logs. Caching and query limits keep those investigations from putting unnecessary load on production systems.
Add to Delivery
I set the Checkout architecture and led execution across more than 20 engineers in four organizations. My part connected delivery data to checkout services and the customer-facing experience, making it possible to offer the flow across shopping pages. I later led its expansion to five European marketplaces.
I was the technical lead for a migration to Java microservices involving more than 30 engineers. I designed core APIs and plugin interfaces and coordinated the rollout. Teams elsewhere at Amazon had to keep shipping features while we replaced the platform they depended on.
Earlier in the migration, I built the framework we used to measure checkout latency. Other work included chaos testing, one-click purchase interactions, and internationalization.
At home
I have a computer that runs research/work agents overnight and emails me a report in the morning. I’ve been experimenting with how to give them useful questions and check their answers.
I use AI coding tools throughout my own projects (CC Fable + Codex Astra combos recently!), including a small collection of systems for notes, studying, biology/genetics, finance and travel.
Other interests
I read about the Bronze Age, especially Mesopotamia and Mycenaean Greece. In 2025 I went to Malta for the temples, which were built before writing reached the island. This October it is Mycenae and Tiryns.
I travel to see paintings. Bosch, Doré’s engravings, Guernica, Mesopotamian reliefs. The Met, the Prado, the Louvre, and the Museo Nacional de Antropología in Mexico City.
A few books I recommend
Energy and Civilization, Vaclav Smil
The Better Angels of Our Nature, Steven Pinker
Antifragile, Nassim Nicholas Taleb
The Scapegoat, René Girard
Use of Weapons, Iain M. Banks
Carrion Comfort, Dan Simmons
Between Two Fires, Christopher Buehlman
1177 B.C., Eric H. Cline
I have rated 348 films. Twenty-two got five stars: 2001: A Space Odyssey, Dr. Strangelove, Fantastic Planet, Blade Runner, Raiders of the Lost Ark, Akira, Ghost in the Shell, Princess Mononoke, Hercules, The Mummy, Requiem for a Dream, Spirited Away, The Fellowship of the Ring, The Witch, Terrified, Hereditary, Chernobyl, The Green Knight, Dune, Pearl, When Evil Lurks, Furiosa.