A company in Salt Lake City that searches for an AI engineer will find, on this page, one who is in Monterrey, Mexico. The city is in the Mountain time zone and is known for software, financial services, healthcare and life sciences, logistics and outdoor products. Each of those asks something different of AI, and the answer I can give is stronger for some than for others.
Mountain time and Monterrey
From early November to mid March Salt Lake City is one hour behind Monterrey. From mid March to early November the two are on the same hour. A call at ten in your morning is at ten or at eleven in mine, so nothing about the day has to be arranged around the distance. The code is reviewed in your own repositories.
Software: a measure before the next feature
Software companies tend to arrive with a feature that already uses a language model and no way to tell whether the next change makes it better or worse. The first thing I build is that measure: a set of tests that decides whether a version ships. The service is LLM evaluation. The public example is the evaluation gates of a multi-agent system that serves a higher-education group with 700,000+ students.
Financial services: access that leaves a record
A financial company asks a different first question: what can the agent reach, and what record remains. No case study on this site comes from financial services. What transfers is governed access to corporate data through an MCP gateway on Cloud Run, and the guide on agent auditability. Enterprise AI is the service that covers it.
Healthcare and life sciences: answers from approved documents
In a hospital or a laboratory, an answer that cannot be tied to an approved document is not usable. No case study on this site is about an AI system in healthcare or life sciences. What transfers is the method for answering from a controlled corpus, under RAG and data platforms, with the same kind of tests in front of every release.
Logistics and outdoor products: quotes, orders and inboxes
A carrier and a company that makes outdoor products have a daily load in common: requests for a price, orders to confirm, email to answer. The published work closest to it is a set of agents around a quoting and deal-tracking system for small and medium companies in Latin America. There the agent drafts the quote, a person keeps the pricing judgment, and each step leaves a trace for review. No case study on this site comes from logistics or from outdoor products. The services are AI agent development and email automation with AI.
Beginning, and the limit
Tell me the role or the problem through the contact page. I read how the process works today and reply in writing: the first thing I would build, the access I would need, and what I would leave unpromised. That can become a contract under my name or a full-time senior role. It does not include anyone on site in Salt Lake City. For work that needs that, I am not the engineer to call.