FixITNow Project Report
1. The Concept
FixITNow is a lightweight web app/PWA concept for booking trusted local handymen for fast home repairs in Phnom Penh. The core idea is a two-sided marketplace that connects renters and homeowners with vetted independent technicians for common issues like plumbing leaks, appliance repair, electrical fixes, and furniture assembly. I chose this concept because it solves a clear everyday problem: when something breaks at home, customers want speed, trust, transparent pricing, and simple communication. The product hook is "Fix it today" with upfront estimates, photo-based issue reporting, and a fast technician match.
2. The Stack
The Stack: HTML5, Vanilla CSS, JavaScript, Web App Manifest, and Service Worker. The AI Assist: ChatGPT helped shape the initial product strategy, target user, pricing model, and launch focus. OpenAI Codex was then used for the development phase, turning the planning notes and UI references into a functional static PWA prototype with reusable CSS, responsive layout, app screens, booking flow, and generated visual assets.
3. The 1 Day Sprint:
Morning (Hours 1-3): We started with product planning. The first decisions were to launch small in one city, focus on Phnom Penh, and target renters/homeowners before expanding to landlords or property managers. We also defined the trust model: vetted independent contractors, fixed price estimates before booking, and quote adjustments only when the technician confirms a more complex issue. Afternoon (Hours 4-6): The coding phase began. Codex built the first PWA structure with a home screen, service categories, estimate card, request form, booking screens, and bottom navigation. The biggest design effort was matching the provided premium mobile UI reference: soft blue gradients, glass cards, large hero typography, technician artwork, realistic icon assets, and a sticky booking flow. The Finish Line (Hour 7-8): The final stretch focused on QA and polish. We checked screenshots, fixed hidden/overflow areas, improved the hero and navigation, replaced weak icon styling with PNG assets, repaired the broken sticky "Review Match" hover behavior, and wrote project documentation. The app was kept simple and static so it could run quickly from a local server without a backend.
4. The Roadblocks
Roadblocks: The hardest part was visual precision. The first versions had layout issues: some buttons stretched too wide, the bottom CTA looked cramped, the hero image was cropped awkwardly, and the navbar did not feel premium enough. Small CSS changes also created side effects, such as the sticky "Review Match" button losing its center position on hover because the hover transform overwrote its fixed-position transform.
5. Key Takeaways
What worked best? The product direction and static PWA approach. FixITNow became clear very quickly because the user problem is obvious, the target market is focused, and the first services are easy to understand. Vanilla HTML/CSS/JS also made iteration fast without waiting on a heavy framework. What worked worst? Pixel-level UI matching and responsive polish. The design required many passes because small details like hero image position, card spacing, button width, icon quality, and sticky hover behavior made a big difference in whether the app felt premium or unfinished.
My final thought: If I had one more day, I would add a real backend mock with technician availability, booking persistence, photo upload previews, and a cleaner admin/technician dashboard. If I could redo the sprint, I would lock the design system first: exact spacing, typography scale, card radius, shadow tokens, and image crop rules before building all screens.