B2B2C Conversational AI @Brarista
Role
Product Designer
Responsibilities
UX Research
Product Design
Digital/Web Design
Product Marketing
Team
Product Team
Tools
Figma
Framer
Adobe Suite
Canva
BACKGROUND
My Time at Brarista 🌱
Brarista is the world's first conversational AI platform built specifically for lingerie retail. By combining intelligent fit technology with professional fitting expertise, it tackles one of the industry’s most persistent and costly problems: 80% of women wear an ill-fitting bra. As a B2B2C white-label solution, Brarista helps lingerie retailers reduce returns, increase shopper confidence, and improve overall profitability.
I first joined as a student consultant, right as the company pivoted toward its conversational AI platform. Today, as a full-time Product Designer, I work across UI/UX design, digital design, and brand storytelling, collaborating closely with our founder, engineers, and a senior designer to shape a product that understands both its end users and the wider lingerie ecosystem.
This case study reflects a year of interconnected work: designing the conversational AI experience, navigating cross-functional collaboration, and building the narratives that define and differentiate the brand.
RESEARCH
Researching With, Not On, People 👯♀️
When I first joined Brarista as a student consultant, my first month was spent conducting a structural study of the ecosystem. Our team dissected the lingerie market, evaluated competitors, analysed consumer behaviours, assessed the product’s UX, and translated everything into a go-to-market plan. Working across this full arc gave me a systems-level understanding of the problem space and grounded every design decision I made afterwards. I learned that good design only happens when you understand the world it’s trying to live in, and this is an approach I’ve carried forward into every practice since.
Building on this foundation, my work transitioned naturally into Brarista’s wider culture of co-creation. As a B2B2C product, we needed to understand retailers and end-users in tandem, and our research practices adapted accordingly. I supported ongoing studies both online and in person, beginning with semi-structured interviews and moving into iterative usability testing as the beta matured.
Group usability session with beta testers
We brought users in at least once a month, with each round of findings thematically analysed and shared with the senior designer and our founder. These insights shaped feature prioritisation and guided our agile sprint cycles. The goal wasn’t just to validate major features, but also to surface friction early and refine the beta through small adjustments, from copy tone to subtle interaction bugs.
Recruitment was “startup-scrappy”: Instagram outreach to our most engaged followers, and Matchable volunteers mirroring our core demographic of working women. We also ran early experiments with Tellet, an AI-driven interview tool that made larger rounds scalable through adaptive questioning.
On a personal level, these conversations stretched far beyond research. Hearing from bra-wearers across ages and identities—young adults, new mothers, cross-dressers—was both educational and grounding. It revealed how culturally charged this product space is, and their stories shaped not only product decisions but also the brand storytelling I later helped develop.
Key takeaways from research:
Co-creation thrives when relationships last
I quickly realised how difficult it is to recruit specialised participants willing to test a product for free. Retaining testers as ongoing co-creators became essential. Users who felt a sense of relationship were far more comfortable giving honest feedback and returning for future sessions. While keeping questions open and unbiased, we still built enough trust and empathy for that relationship to grow over time.
Not every insight deserves equal weight
Working in a fast-paced, resource-tight environment taught me to distinguish deal-breakers from “nice-to-haves.” Weighing user value against development cost became part of my design judgement, helping focus our efforts where they mattered most.
DESIGN
Designnnnnnn… (ongoing)
Research surfaced the core design challenge at Brarista: building a lingerie shopping assistant that could live quietly inside each retailer’s site while still earning end-user trust in a category where expertise is deeply human. The design had to adapt to every brand environment it entered without dissolving its own logic, which was a new design problem for me.
Three forces shaped what the product could realistically be:
Model behaviour: what the system could reliably understand and respond to.
Retailer ecosystems: the aesthetic and technical rules of Shopify, Wix, BigCommerce, and each client site.
Interface scale: a small embedded widget/pop-up.
Within these constraints, my work moved across three intertwined layers:
Interfaces
Mobile and desktop chatbot interfaces, plus the embedded Brarista fitting quiz. I’m now also exploring the next set of conversational AI touchpoints.
Conversational UX
Prompt engineering and behavioural QA alongside the founder and engineers.
Systems
Early groundwork for a design system, finding a balance between internal consistency and each client’s brand environment. Still very much in development.
Below are product interfaces I actively contributed to throughout the year. You can also see them customised and live on our retail clients’ sites, listed on Brarista’s website. The screens that follow are arranged in the order a user experiences the flow.
Entry Point Redesign
Role: Led by senior designer. I refined UX details and aligned micro-interactions with developers.
Chatbot Landing Page Redesign
Role: Co-designed. Early exploration by senior designer; I completed the UX and brought it into production.
The Brarista Fitting Quiz
Role: Led by senior designer and founder. I contributed to ideation, UX prototyping, and collaborative decision-making. Still evolving.
Our design rhythm was conversational; ideas surfaced in passing chats or during debugging. I learned to capture them and revisit them a week later, keeping the process generative while filtering for what actually mattered.
Next Steps
I’m leading the design of new product touchpoints and contributing to an expanded fitting quiz. As the product matures and the ecosystem widens, system-level work (standardisation without losing cross-retailer adaptability) is becoming a core part of my role. This will become its own case study once complete, and I’m looking forward to carrying that clarity into the next iteration of the product!
STORYTELLING
Making Bra Fitting Culturally Relevant and Cool 😎
As the product matured, we shaped Brarista’s narrative alongside it. The interface and the story developed together; understanding one sharpened decisions in the other.
A large part of my work centred on how potential retail clients first encountered the brand. I designed the “Wall of Love” and “Brarista vs. Other Tools” pages end-to-end, both structured to address objections clearly and build credibility through evidence and social proof. I also wrote and designed customer stories and blog content that framed Brarista’s milestones and industry perspectives in a way retailers could immediately understand.
This narrative work naturally expanded onto social platforms. I managed both the LinkedIn and Instagram presence—LinkedIn as our B2B home, Instagram as the space where decision-makers spent their off-hours. As a B2B2C product, our content needed to speak to retailers while also engaging end consumers. A core strategy was positioning Brarista (and Bella as its human voice) as a contemporary, authoritative figure in the lingerie space. I created editorial-style posts, industry insights, and more playful cultural takes that framed lingerie and bra fitting as something relevant, empowering, and sometimes just fun or sexy. The goal was to make lingerie a “cool” topic, and to place Brarista at the centre of that cultural shift.
Working closely with our optimisation and analytics lead also gave me exposure to technical marketing. I learned how content structure influences search visibility, and how analytics reveal what users pay attention to. This shaped how I structured website pages, social content, and built narratives, keeping the storytelling grounded in real behaviour.