B2C / B2B Web App
Web App Design
Product UI & UX
HousePik AI transforms the overwhelmed home-buying decision into a personalized, data-driven choice. I didn't just design this. I researched it, architected it, built it, and shipped it live in production, end-to-end.
Industry:
Prop-tech, AI
Project Status:
Live
Timeline:
Nov 2025 - Jun 2026
My Role
Solo — Product Designer + Design Engineer
Project type
B2C / B2B Web App
Tech stack (prototype)
Figma · Next.js 16 · TypeScript · Tailwind CSS v4 · Supabase · Vercel AI SDK · OpenRouter / DeepSeek
Designes and builtof an AI-first chat experience that converts user-submitted property data into a defensible, explainable shortlist for reducing decision friction and enabling fast market validation.
I didn't just design this. I researched it, architected it, built it, and shipped it — live in production, end-to-end.
The Problem
Decision paralysis is the last mile nobody solved
Buying or renting a home is one of the most financially loaded, emotionally charged decisions a person makes. And after all the tours, the spreadsheets, and the exhausted conversations with a partner most people are still stuck.
Not because they lack information. Because they have too much of it, with no framework for what they actually prioritize.
Current tools make this worse, not better:
Zillow, Redfin, and Realtor.com surface more options but offer no help actually choosing
Generic scoring apps rank properties using someone else's weights, not yours
Couples argue because their priorities are never quantified only assumed
Agents guess at what clients actually want versus what they say
Users revisit the same 3 properties 4 times and still can't commit.
Quote pulled from research insight
What I shipped
A guest-first, conversational onboarding chat that collects just-enough data (≤3 required fields) and progressively reveals optional criteria.
A one-click property ingestion flow (paste link / upload photos / quick add) with parsing + confidence feedback.
Side-panel Results Analysis: ranked “Best Choice #1 / #2 / #3”, confidence indicators, short “Why this fits you” explanations, and “Immediate Recommendations” actions.
Project model with three tabs: Summary, Properties, Collaborate, with collaborator invites gated to signed-in users (owner-controlled permissions).
Single guest analysis limit + email-save (one-time secure link) to drive sign-ups without blocking initial value.
Live prototype (AI no-code) used for user testing & market validation
The real insight
The Strategic Position
HousePik AI is the final step — the decision layer.
That's an entirely unoccupied position in the real estate tooling market.
My Role
TSolo: Design + Engineering, end-to-end. The line between designer and engineer collapsed completely.
Product & Strategy
Problem framing
Market positioning
Persona definition
Feature prioritization
Design
User research
Information architecture
UI system & components
Interaction design
Engineering
Next.js frontend
Vercel SDK
Supabase database
Production deployment
My role & responsibilities
Led product & UX strategy, user journey design, and interaction design.
Owned IA, onboarding, and analysis UX patterns.
Built the AI no-code prototype for live user testing.
Prioritized features and trade-offs with product & engineering stakeholders.
Designed collaborator flows & permission model.
Ran early usability sessions and iterated flows.
Users & audience
Primary: First-time home buyers and couples who need collaborative decisions
Secondary: Repeat buyers who want speed and confidence.
Use case: compare 3–6 candidate properties and choose a best fit quickly.
The solution: what the product does (core flows)
Next steps & growth roadmap
How it works in 30s
Conclusion
The HousePik design concept proves that solving home-buying paralysis isn't about more data, it’s about better clarity. By moving away from rigid forms and toward a "value-first" conversational model, we’ve created a roadmap for a product that earns user trust before asking for a sign-up. This project stands as a strategic foundation for testing how explainable AI can transform a stressful, high-stakes search into a confident, collaborative decision.
Scroll Up