I shape how systems
behave so Humans
feel heard.


About

late-night-learner,
reformed-screenwriter,
IVF-Mumma,
worth-the-climb immigrant,
half-of-a-marriage,
occasional-late-night-dancer,
and an out and proud verbivore.

I design the language, context, and conversational structures that shape how intelligent systems behave. My work is less about scripting responses and more about understanding why interactions succeed or fail, reducing ambiguity, designing for edge cases, and helping AI behave in ways that feel reliable, natural, and genuinely useful.


I came to conversation design from screenwriting. Turns out, building high-stakes dialogue for characters and designing natural interactions for AI systems require the same core skill: understanding subtext, mapping human intent, and knowing exactly what a user needs to hear (and when) to move an interaction forward.


Using tools like Figma, Notion, Amazon Connect, and OpenAI Playground, I treat language as part of the system architecture. I prototype user flows, iron out edge cases, and design the conversational logic that transforms cold, mechanical system prompts into intuitive, human-centered experiences.  

What keeps me curious?


I'm fascinated by the gap between what an AI system is capable of and how it actually behaves in the real world. Most interaction failures aren't model failures, they're failures of context, instruction, ambiguity, or evaluation. I enjoy finding those gaps and designing systems that make intelligent behaviour more reliable.
Case Studies

The Human Side of IVR

Behavioral Health / IVR Design

Amazon Connect + Amazon Lex
2025
Portfolio Case Study

Designing conversational systems where safety
overrides certainty.

The Starting Point

I've navigated the medical system during high-stakes moments. I know what it feels like to need support and find it, and what it feels like to need support and not find it. That experience is why behavioral health felt like the right context to learn IVR design in, not just a technically interesting problem, but a human one.

This sample demonstrates how to engineer clear, conversational guardrails within a hybrid deterministic/NLU architecture using Amazon Connect and Amazon Lex.


Read Full Case Study
 

The "Who Gets the Aux?" Solution

Conversational AI / Playlist Design

Figma / Conversational Design
2025
Solo Design Exploration

Designing AI mediation for competing user intent.

Why I Built This


I've lived in shared spaces my whole life, family homes, share houses, roommates, partners. And I've noticed that the small things matter more than they should. Who controls the music is never really about the music. It's about whether your presence in the space is acknowledged.


Read Full
Case Study

Ending the "Empty Box" Anxiety

Conversational AI / Clarification Design

Figma (screen design), Notion (dialogue design)
2026
Solo Design Exploration

Designing AI behaviour under ambiguity.

It Started With

When my brain is messy, the last thing I need is to be asked to organize it first. Thoughts feel fragile and fleeting, if I have to sort them before I can say them, I'll lose them. 

The design challenge wasn't helping people write better, it was helping an AI system understand unstructured intent without forcing users to organize their thoughts first.


Read Full Case Stu
dy


Icebreaker Jake

Prompt Engineering 
/ Conversational AI

OpenAI Playground, Figma, Whimsical
2025
Solo Design Exploration


The Question I started with:

What happens when you treat a system prompt as a design material, and keep iterating until the outputs stop sounding like LinkedIn and start sounding like a friend that’s next to you on the couch?


Read Full Case Study





© Laura Bray 2026