page
About Me
Background, perspective, expertise, and interests.
I’m a technologist and builder with more than 25 years of experience designing, building, and evolving software systems.
Any questions?
My work has spanned software development, enterprise architecture, cloud computing, data, and artificial intelligence across large and complex organizations.
Today, I’m particularly interested in what happens to software architecture and engineering as AI becomes an active participant in how systems are designed, built, tested, operated, and evolved.
What I’m exploring
AI can generate software faster than we’ve ever been able to produce it.
That makes the questions surrounding the code increasingly important:
- What should we build?
- What constraints should shape the system?
- Which quality attributes actually matter?
- How do we evaluate architectural tradeoffs?
- How do humans and AI agents work together effectively?
- How do we move from impressive prototypes to durable production systems?
- How do we keep increasingly complex systems understandable?
These aren’t entirely new questions.
Many of the disciplines developed over decades of software engineering remain highly relevant: software architecture, quality attributes, architecture decision records, architectural evaluation, domain modeling, distributed systems, cloud architecture, and operational excellence.
What’s changing is how we apply them.
Architecture in the age of AI
I’m exploring the intersection between established software architecture practices and emerging AI-native engineering approaches.
That includes:
- software and enterprise architecture
- generative AI and agentic systems
- context and specification engineering
- architecture decision records
- quality attribute workshops and ATAM
- C4 and architectural modeling
- cloud and distributed systems
- AI evaluation and governance
- developer tools and AI-assisted engineering
- human + agent engineering workflows
I’m especially interested in how these practices can become more executable.
Instead of architecture existing primarily as documents and diagrams, what happens when architectural knowledge becomes context that AI agents can reason about, evaluate against, and help maintain?
That’s an area I expect to spend considerable time exploring.
Building, not just theorizing
Architecture works best when it remains connected to implementation.
I continue to build software, experiment with AI models and agents, work with cloud infrastructure, test emerging developer tools, and create small systems to explore ideas.
Those experiments often become the material for what I write about here.
Some validate an idea.
Some expose flaws in it.
Both are valuable.
Professional background
I’ve spent more than two decades working across enterprise technology in industries including financial services, healthcare, cloud computing, and insurance.
My career has progressed from software development into application, solution, cloud, and enterprise architecture, along with technical and architecture leadership.
I’ve worked inside large enterprises, served as a Solutions Architect at Amazon Web Services, founded a healthcare analytics technology company, and built independent software and technology projects.
That experience has given me a perspective that spans implementation, architecture, organizational constraints, and technology strategy.
Learning in public
PhillipJMurphy.com is where I document what I’m learning, building, and thinking about.
You’ll find architecture notes, experiments, technical explorations, demonstrations, and longer-form thinking about AI, software, cloud computing, and the changing practice of technology.
I’m less interested in predicting exactly where AI is going than in experimenting with what’s possible, understanding the constraints, and documenting what I learn along the way.
The goal is simple:
Learn. Build. Understand. Share.