Embedded Files
“The best way to understand the future is to build small pieces of it.”
“The best way to understand the future is to build small pieces of it.”
“I had an idea on Tuesday” and “I had a working version by Friday” used to be two very different sentences. AI has changed that.
This page is a running portfolio of the tools, prototypes, and experiments I’ve built with AI. It's what shipped, what flopped, what surprised me, and what I learned along the way.
I’m especially interested in how AI changes the product development cycle: how quickly we can move from problem → prototype → user feedback, and how much more room that creates for curiosity, testing, and better decisions.
These projects are not just about building faster. They’re about learning faster.
A lifecycle growth prototype for adventure travel, built to show how customer intent can drive repeat bookings, referrals, reviews, and first-trip conversion.
Problem spotted: Lifecycle growth is often treated as a messaging problem: send better emails, improve the welcome flow, recover abandoned carts, ask for reviews. But in adventure travel, the real opportunity is deeper. Customers are not just buying dates and destinations. They are buying a feeling: reset, challenge, connection, escape, celebration, or the chance to feel brave again. A strong lifecycle system should understand those motivations and use them to guide the right next message, trip, channel, and experiment.
What I built: Adventure Lifecycle OS is an AI-enabled lifecycle growth prototype designed for an adventure travel marketplace. It maps the customer journey from visitor to signup, trip browsing, first booking, post-trip engagement, repeat booking, referral, and review. The prototype includes a lifecycle journey map, customer segments, a campaign studio, an insights dashboard, and an experiment backlog. Each section is built as a practical artifact that could be used in a planning session, not just as a conceptual mockup.
The product angle: The prototype turns customer intent, trip behavior, and lifecycle stage into targeted growth opportunities. It includes eight working customer segments, a nine-stage adventure-travel lifecycle, multi-channel campaign planning across email, SMS, WhatsApp, and in-product moments, plus mock north-star metrics for repeat booking rate, referral rate, signup-to-first-booking conversion, and long-term LTV.
AI’s role: AI helped move quickly from job brief to product strategy, journey model, segmentation framework, campaign logic, copy examples, and dashboard structure. The goal was not to create generic lifecycle messages. It was to show how AI can help build the operating system around lifecycle growth: identifying moments, structuring experiments, translating user signals into campaigns, and making strategy visible.
What I learned: The most useful AI-enabled growth tools do not just generate copy. They help teams make better decisions about timing, segmentation, prioritization, and customer motivation. In adventure travel, especially, emotional intent can be a powerful signal. Someone looking for “a reset” may need a very different lifecycle path from someone looking for “a bigger challenge,” even if both are browsing similar trips.
Why it matters: This project shows how I would approach a lifecycle growth role: by connecting product thinking, customer psychology, CRM strategy, experimentation, and AI-assisted execution. It demonstrates that I can take a company’s growth priorities and turn them into a usable system for increasing repeat bookings, referrals, reviews, signup-to-first-booking conversion, and long-term customer value.
Built with: Lovable, ChatGPT, Claude, AI-assisted product strategy, lifecycle journey mapping, customer segmentation, CRM campaign planning, and experiment design.
See it here: https://adventure-compass-lifecycle.lovable.app/
Product insight:
Product insight:
The quiz is the front door. The insights dashboard is the business value. Together, they show how a simple user experience can create a feedback loop between customer intent, personalization, and brand strategy.
Problem spotted: Most travel discovery tools start with logistics: dates, budget, destination, group size. But people often want to travel because of an emotional need: burnout, restlessness, reconnection, novelty, grief, celebration, escape, or the desire to feel more alive. Traditional search captures what people type. It rarely captures why they are searching.
What I built: A trip-matching experience where users describe what they are craving emotionally, and the tool suggests the shape of an escape that might fit. Instead of jumping straight to destinations, it starts with intent: “I need room to breathe,” “I want to feel properly alive,” “I’d love to share it with good people,” or “anything but the usual.”
The second layer: I also built an insights view that shows the kind of data a travel brand could collect from this experience: emotional drivers, preferred trip styles, group context, escape motivations, and patterns in what people are really seeking. That turns the quiz from a lightweight discovery tool into a potential research, personalization, and demand-sensing layer.
AI’s role: AI helped turn a fuzzy emotional insight into a structured product flow, matching logic, UX copy, prototype screens, and insight categories. It also helped explore how qualitative feelings could become useful product and marketing signals without making the experience feel clinical.
What I learned: Emotional intent is incredibly useful product data when it is gathered in a way that feels natural to the user. “I need to switch off,” “I want novelty,” and “I want to reconnect with someone” are very different travel briefs, even if all three users search for “weekend break.”
Why it matters: This kind of tool could help travel companies improve personalization, content strategy, email segmentation, campaign messaging, package recommendations, and product development. It gives brands a way to understand not just where people want to go, but what emotional job they need the trip to do.
Built with: Claude and Lovable in 3 days
User tool: https://your-kind-of-adventure.lovable.app
Insights view: https://your-kind-of-adventure.lovable.app/insights
A custom planning app for a five-week trip through New Zealand, Australia, and Singapore.
Problem spotted: Every trip planner I tried handled about 60% of what I needed. The rest ended up in spreadsheets, notes apps, email folders, currency converters, and a growing sense that “surely this should be easier.”
What I built: A single trip command center for flights, accommodation, activities, transport, packing, weather, maps, and shared expenses. It included receipt parsing, categorized spend tracking across multiple currencies, and a color-coded calendar that made the whole trip visible at a glance.
AI’s role: AI helped me move quickly from messy personal requirements to a working product structure, data model, UX flows, and build-ready prompts. Instead of spending weeks documenting the perfect spec, I could prototype, test, and improve the tool while the trip was still taking shape.
What happened: We used it daily for five weeks. It survived flight changes, surprise tour add-ons, three currencies, and the very real test of using it while tired, hungry, and standing in an airport. The receipt parser became the feature we leaned on hardest, because no one wants to manually enter expenses after dinner.
Product lesson: The best travel tools are not just itinerary tools. They reduce cognitive load. They help people feel oriented, in control, and less likely to argue over whether 25°C is hot or cold.
Built with: Claude and Lovable in 3 days
See it here: https://wanderlust-palooza.lovable.app/
A website critique tool that reviews your site like a seasoned product manager would.
Problem spotted: Most website audits are either too technical, too generic, or too narrow. They might flag SEO issues, page speed, or accessibility concerns, but they often miss the bigger product questions: Is the site clearly positioned? Does it explain the value quickly? Where are users likely to hesitate? What opportunities are being left on the table? And increasingly, is the site readable not just to humans, but to AI-powered search and answer engines?
What I built: ProductLens is an AI-powered website critique tool for founders, marketers, and product teams who want an external perspective on their site, fast. The tool scrapes a website, reads it through a product strategy lens, and critiques it across five dimensions: strategic positioning, friction points, prioritized recommendations, AI search readiness, and untapped opportunities.
The scoring layer: One of the key features is an AI search readiness score, rated from 1 to 10. This helps users understand how well their site communicates with emerging AI search and discovery systems, not just traditional search engines. The goal is to make AI visibility feel practical and understandable rather than vague or intimidating.
AI’s role: AI powers the site review, but the value comes from the structure around it. I designed ProductLens to mimic the kind of critique a seasoned product manager might give: clear, direct, commercially aware, and focused on what would make the site easier to understand, trust, and act on.
What I learned: The most useful AI products do not just generate output. They apply judgment through a clear framework. For ProductLens, the challenge was turning a subjective review process into something repeatable, actionable, and useful to different types of users, from solo founders to product and marketing teams.
Why it matters: As discovery shifts, websites need to work harder. They need to communicate clearly to potential customers, search engines, AI tools, and decision-makers who may only give a page a few seconds before moving on. ProductLens explores how AI can help teams identify weak spots, sharpen positioning, and uncover opportunities without waiting weeks for a full external audit.
Built with: AI-assisted product strategy, Lovable, website scraping, structured audit frameworks, and my product management/AI SEO review process.
See it here: https://productlensinsights.app/
A digital trivia platform that turns fun fact books into playable quiz experiences.
Problem spotted: My fun fact books already had the ingredients for great trivia nights: surprising facts, quiz questions, themed topics, and readers who enjoy testing what they know. But the experience lived mostly on the page. I wanted to explore what would happen if the books could become interactive, replayable, and shareable.
What I built: A digital quiz platform where readers can play themed trivia based on my books and printable quiz packs. The site supports solo play, host-led group play, room-code style quiz sessions, quiz detail pages, coming-soon quiz listings, and a growing library of themed quizzes including New York, Canada, Alaska, Tennessee, Christmas facts, astronauts, and quirky careers.
AI’s role: AI helped accelerate the product structure, UX flows, question-bank formatting, quiz logic, landing page copy, SEO metadata, schema recommendations, and Lovable build prompts. It also helped turn a large amount of book content into organized, playable quiz experiences without losing the Knowledge Nugget tone.
What I learned: AI is especially powerful when paired with existing intellectual property. The hard part was not generating trivia questions. The hard part was building a product system around them: difficulty levels, answer explanations, quiz formats, replay value, content quality, and a clear reason for readers to move from book → printable → digital quiz.
Why it matters: This project shows how AI can help creators extend existing content into new product formats. It is part publishing experiment, part edutainment platform, part product ecosystem. It also demonstrates how AI can speed up the path from static content to interactive digital experience.
Built with: Lovable, Claude, ChatGPT, existing Knowledge Nugget book content, and a lot of product wrangling.
See it here: KnowledgeNuggetQuizzes.com
Page updated
Google Sites
Report abuse