Proof of Work
A working record — one flagship product, and the experiments alongside it.
Founder · Consumer AI Product
BECOME
Human-centred AI designed to help people understand who they're becoming.
View Case Study
Government Transformation
AI for Public Services
Agentic AI concepts supporting citizens through complex application journeys while reducing manual verification for officers through human-in-the-loop workflows.
Explore the Approach
Workflow Automation
AI Appointment Assistant
Conversational booking integrated with calendars and automated reminders.
See Details
Enterprise AI
Conversational AI & Digital Assistants
Chatbots and internal assistants across facility management, insurance, customer service, and e-commerce.
See Details
Product Discovery
Rapid AI Innovation
Rapidly designed and validated AI prototypes exploring document intelligence, citizen assistants, workflow automation, and decision support for public-sector services.
See Details
Flagship Case Study
BECOME
Product Hero
Overview & Role
An AI journal that helps people see who they are becoming while they are still living it. Founder, sole designer, and sole builder — from spec to shipped product.
The Challenge
People are changing constantly, but rarely see it while it's happening. Journals capture what happened, not what it means.
User Insight
The Conscious Seeker: late 20s to 40s, already doing some version of self-reflection, but missing a single place to make sense of it over time. Not a beginner at self-work — missing the mirror.
Product Principles
- Quiet by default — no clutter, no dashboards on empty states.
- Mobile-first — designed for the phone at the end of a long day.
- Data visualisations only when there's data worth showing.
PRD to Live
PRD Written
Built Nights & Weekends
Founder-Led Personal Use
Live
3 weeks from PRD
Product Walkthrough
Onboarding opens with the founder's own story, then straight into a first journal prompt. Write freely, BECOME reflects inline. After 5 entries, patterns surface. After 10, strengths and blind spots are framed gently. After 4 weeks, the Inner Compass becomes a living document.
AI Behaviour & Guardrails
The AI doesn't give advice or direction. It's designed as a wise, close friend who has been reading your journal for years — asking the question that unlocks something, and trusting you to find your own answer.
Deliberately Excluded
- No social features — this isn't a place to perform reflection.
- No habit tracking or productivity mechanics.
- No onboarding quiz — the AI learns from what you write.
- No therapist integrations — BECOME is not therapy.
Screens & Interactions
Product System
Journal Entry
Next.js
AI Reflection
Claude API
Data & Auth
Supabase
Analytics
PostHog
Testing & Iteration
Founder-led personal testing happened before anyone else touched the product — including two days of after-hours debugging on the Inner Compass feature that ended with reading what it had written, and crying, because it was accurate.
Results & Current Goals
Near-term goal: 100 active users with weekly retention above 40%.
Lessons for Client Work
Every decision in BECOME — down to what the AI is not allowed to do — is a decision I now bring into client work: what makes people trust an AI, and what makes them abandon it.
Roadmap
A weekly reflection digest, guided prompts for specific life chapters, exporting your journey as a document, and eventually sharing entries with a trusted coach, therapist, or partner.
Government Transformation
AI for Public Services
- Challenge
- High-volume government services often require significant manual verification and repetitive citizen support.
- Approach
- Designed AI-assisted citizen journeys and multi-agent workflows that support document verification, decision support, and human-in-the-loop processes.
- Impact
- Showed how practical AI can improve service delivery while reducing administrative effort, with all project details intentionally anonymised.
Builder Experiments
Rapid AI Prototypes
From appointment assistants to workflow automation — practical AI concepts built and tested quickly.
AI Appointment Assistant
- Challenge
- I prefer booking appointments by text, but my regular nail and hair salons still relied on phone calls and manually updating appointment books. The process worked, but it created unnecessary back-and-forth for both customers and staff.
- Approach
- I designed and built an AI-powered booking assistant that lets customers book through chat. The assistant checks availability against Google Calendar, creates appointments automatically, and sends confirmation and reminder messages before each booking.
- Impact
- The prototype demonstrated how a familiar messaging experience could replace manual scheduling, reduce administrative effort, and create a smoother customer experience without changing how the business operates.
Conversational AI & Digital Assistants
- Challenge
- Whether helping residents report maintenance issues or helping customers find products, many everyday interactions were repetitive, manual and difficult to scale.
- Approach
- Over the past decade, I've designed chatbot and internal assistant experiences for facility management, insurance, customer retention and e-commerce, focusing on solving practical business problems rather than simply adding another chat interface.
- Impact
- Each solution improved access to services, automated repetitive interactions, and became part of broader digital transformation initiatives within the organisations.
A Few Examples
- Facility Management Assistant — maintenance requests, air-conditioning servicing, booking technicians and service updates.
- Customer Retention Chatbot — guided users through churn-saving journeys and personalised retention offers.
- Smart Product Assistant — helped customers discover products and stock availability across e-commerce platforms.
- Internal Service Assistant — streamlined employee support and common operational requests.
Rapid AI Innovation
- Challenge
- Public-sector teams need practical ways to explore AI before investing in full implementation.
- Approach
- Rapidly designed and built AI prototypes focused on citizen services, document intelligence, workflow automation, and internal decision support.
- Impact
- Validated product concepts quickly and demonstrated how human-centred AI could improve service delivery and operational efficiency.