← eliuthchaverojasso.com

Writing

Notes from the field.

Short, first-hand notes about building software where AI, engineering workflows and physical systems meet.

In progress

I am developing these notes from actual product and engineering work. They will go live when there is a finished argument, a reproducible artifact or a lesson worth making public.

Engineering AI

Why LLMs Shouldn't Be the Final Authority in Engineering Verification

What changes when an AI system has to support a decision someone must review and defend.

Forward-deployed engineering

What Building AI for Engineers Taught Me About Forward-Deployed Engineering

Why direct exposure to the domain changes architecture, product definition and delivery.

Revit

Five Production Failures That Change How You Architect a Revit Add-in

Host lifecycle, transactions, versioning and model state as product constraints.

Evidence

Building AI Systems That Can Show Their Work

Evidence provenance, counterexamples and the boundary between investigation and determination.