Built an AI Voice Agent for hospital appointment booking using Vapi, Next.js, and NestJS. It can collect patient information, help select the doctor/service/time, and automatically book the appointment directly into the website calendar.
A practical example of how voice AI can simplify scheduling and reduce front-desk workload.
I'd treat the calendar write as the critical boundary here. How do you handle an interrupted call after a slot is reserved so a retry cannot create a duplicate appointment?
The voice agent has the client’s email or phone number. Before making any appointment, it checks the database to see if an appointment already exists for that date.
If an appointment already exists, it doesn’t create a duplicate. It asks the user if they want to update or reschedule the existing appointment.
Designing a fragrance website with 2 sections that feels as premium as the product.
Built a fragrance experience where the product takes center stage.
I wanted the website to feel as refined as the fragrances themselves minimal layout, strong typography, immersive product visuals, and subtle interactions that make exploring each scent feel premium.
The goal wasn’t to fill the page with elements, but to give every product enough space to create its own presence.
Built with Next.js, with a focus on responsive design, smooth interactions, and a polished browsing experience.
The product imagery is doing most of the work here. Did you use art-directed mobile crops, or keep one image with a different object position? That can preserve label detail without loading the full desktop asset.
Building a Modern Restaurant Management Dashboard with Lovable (Light ☀️ vs. Dark 🌙)
I’ve been working on the dashboard interface for ReserveTable (Eat Pro Japan), focusing on giving restaurant managers and users a clean, modern, and intuitive interface to manage bookings, tables, and analytics.
Using Lovable, I was able to rapidly prototype and construct a modern UI system complete with smooth component transitions and native theme switching.
✨ Key Design & Tech Highlights:
⚡ Built with Lovable: Leveraged AI-powered design-to-code workflows to assemble clean component hierarchies, responsive search/filter bars, and sleek cards rapidly.
🌗 Seamless Light & Dark Mode: Designed a flexible color system using modern CSS variables / Tailwind tokens so the entire interface gracefully shifts between an elegant light aesthetic and a moody, dark dining ambience.
🎯 Contextual UI Elements: Styled input controls, select menus, and primary CTA buttons to maintain legibility and contrast across both light and dark themes.
For reservation and table states, the same semantic color tokens need to stay clear in light and dark mode. Did you define status tokens once and map both themes to them, or tune those colors per component?
An AI workflow that reads every incoming email, sorts it into 6 categories, and takes the right action automatically.
The problem: businesses lose time and leads because customer emails are mixed with promotions, job alerts, and security emails.
What I built:
Gmail Trigger that picks up each new email
AI Text Classifier (Google Gemini) with 6 categories: Security Alert, Promotions, Jobs, Personal, Payments, and Client Inquiries
Automatic Gmail labels for every category
Google Sheets log for jobs, leads, and payments
AI information extraction for client emails (name, phone, service, budget)
AI-written reply drafts for client emails, saved as drafts so a person approves every reply
Instant Discord alerts for security, payment, and client emails
Tools: n8n, Google Gemini, Gmail, Google Sheets, Discord
This project is a working demo built on my own inbox. The same system can be adapted to any business email, such as travel agencies, event venues, stores, recruiters, and consultants.