AI Work
Human-AI Interaction & Agentic Systems
I design AI experiences that help people work effectively with intelligent systems — from conversational assistants and agentic workflows to Human-in-the-Loop tools and AI-supported services.
My work focuses on the interaction between people, AI behaviour, workflow logic, and real operational systems: deciding what AI should do, where it needs boundaries, when people should remain in control, and how the experience should recover when the AI gets something wrong.
I bring more than a decade of UX, product, and service-design experience into this work, supported by hands-on experience prototyping with LLMs, agent workflows, APIs, automation platforms, and structured AI systems. That technical fluency lets me move beyond conceptual AI experiences and test how design decisions behave in working prototypes.
The projects below explore different parts of that practice — Human-AI Interaction, Conversational AI, agent behaviour, workflow orchestration, human oversight, AI-specific evaluation, and the integration of AI into existing products and services.
Selected work
Selected AI Projects
R.A.I.N.E.
An AI-assisted intake and qualification workflow designed around explicit boundaries between AI interpretation, deterministic system logic, and human authority. R.A.I.N.E. interprets incoming inquiries, identifies missing context, prepares adaptive follow-up communication, and keeps customer-facing actions behind human review.
Bob’s Bots Scheduling Assistant
A conversational AI scheduling assistant designed to translate natural-language intent into grounded, tool-backed service actions while preserving clarification, confirmation, and user control.
AI UX Research Assistant
A structured AI workflow designed to help transform raw usability research into findings, domain-aware review, and synthesized reporting while preserving human judgment throughout the research process. The project explores specialized agent responsibilities, structured handoffs, Human-in-the-Loop review, and the role of AI within an established professional workflow.
Design principles
My Approach to AI Design
I approach AI as an interaction and systems-design problem, not simply a model-selection or prompt-writing problem.
That means thinking deliberately about:
AI Behaviour
What should the AI understand, generate, recommend, or act on — and what should remain outside its authority?
Human Control
When should people review, correct, override, approve, or take over from the AI?
Conversation
How should the system clarify uncertainty, ask for missing information, confirm consequential actions, and recover when communication breaks down?
Workflow & State
How does the experience move between AI, people, tools, APIs, business rules, and systems of record?
Trust & Recovery
What happens when the AI is uncertain, incomplete, incorrect, or unable to complete the task?
Evaluation
How do we test not only whether an AI response sounds good, but whether the system behaves appropriately across normal, ambiguous, and failure scenarios?
The goal is not to maximize AI autonomy.
It is to design the right relationship between AI capability, deterministic systems, and human judgment.
Implementation fluency
From Design to Working Prototypes
I prototype AI systems closely enough to understand how interaction decisions behave under real implementation constraints.
My work has included tools and technologies such as OpenAI APIs, n8n, MCP, Vapi, LangChain, LangGraph, Python, and TypeScript, alongside the UX and service-design methods I use to understand workflows, user needs, operational constraints, and system behaviour.
Technical implementation is not the end goal.
It gives me a way to test AI behaviour, identify failure modes, collaborate more effectively with engineering teams, and make better product decisions.
Technical literacy makes me a stronger AI designer.
Beyond the portfolio
Exploring AI for your business?
Alongside my product and design work, I use ForwardVantage.ai to explore practical applications of AI agents, workflow automation, and AI-enabled services for real business processes.
If you are looking at where AI could fit into an existing service, internal workflow, or customer experience, you can learn more about that work at ForwardVantage.
Connect
Let’s Talk
I’m interested in opportunities involving Human-AI Interaction, AI Product Design, Conversational AI, Agentic Workflow Design, AI/CX Strategy, and the design of AI-enabled services and products.
If you’re building AI experiences that need to be useful, understandable, controllable, and grounded in real workflows, I’d be happy to connect.
Contact
If you'd like to talk about service design, product design, AI workflows, or a future role, send a note.
