Data-led solutions architecture for systems with consequences.
I am a hands-on data and software engineer developing toward Solutions Architect roles. My work spans data engineering, AI engineering, software engineering, and technical program leadership, with solutions architecture connecting those capabilities from discovery through adoption.
I translate ambiguous workflows into solution boundaries, delivery paths, and production systems that teams can operate and adopt.
I currently work in software and data engineering at Moderna, building production applications and data workflows for scientific teams. Earlier roles at Meta Reality Labs, Google Cloud, Google Shopping / Express and the State of California exposed me to a very different set of systems: hardware programs, global operations, logistics, public platforms and large cross-functional programs.
That range created a bias toward practical solutions architecture. A technically elegant system still fails if it ignores workflow fit, permissions, integration constraints, failure recovery, operating ownership, or the economics of adoption.
Data engineering is my primary recruiting path, supported by AI engineering, software engineering, and technical program leadership. Solutions architecture is the connective practice: understanding a customer or scientific workflow, shaping system and data boundaries, de-risking difficult integrations, and connecting implementation to measurable outcomes.
My technical writing covers production AI, enterprise RAG, agentic systems, evaluation, scientific software, and forward-deployed engineering. I use reference architectures and explicit assumptions rather than presenting hypothetical designs as employer systems. The articles show how I frame decisions; the Experience page shows the professional work behind that judgment.
I also study the business side of technology through MBA coursework at Boston University, building on graduate training in electrical engineering.
For the full career chronology, view my LinkedIn profile ↗
Discover, design, deliver, and earn adoption.
My focus is the complete solution path: technical discovery, solution design, delivery integration, and operational adoption.
Applied AI
Generative AI, agentic AI, RAG, LLM evaluation, context engineering, human-in-the-loop systems and observability.
Software architecture
Distributed systems, APIs, workflow orchestration, state ownership, retries, idempotency, reliability and production operations.
Scientific systems
Biotech and life sciences software, scientific R&D workflows, evidence provenance, experiment context and decision systems.
Forward deployed engineering
Technical discovery, architecture scoping, rapid vertical-slice prototypes, customer deployment and productization.
Tools I’ve used in production work.
I’d rather be specific than exhaustive. These are technologies represented in my professional experience.
Four recurring questions.
How should AI fit into scientific work?
Not “where can we add an LLM?” but where intelligence, retrieval, tools, provenance, evaluation and human judgment create leverage.
How should production AI fail?
Retries, state ownership, idempotency, permission boundaries, partial execution and human escalation are architecture decisions.
What makes a field deployment reusable?
A strong FDE solves the customer problem while identifying the stable primitive that belongs in the platform—and the bespoke detail that does not.
How do technical leaders create leverage?
By making good decisions repeatable: clear architecture boundaries, design reviews, reusable interfaces, migration paths, operating mechanisms, mentoring, and alignment across teams.