Back to Projects
Case Study Β· Marketplace Platform

Gardenly

AI-led Product Development for a UAE Garden Services Marketplace

Role

AI PM + Full Stack + Marketing

Stack

Next.js Β· FastAPI Β· PostgreSQL

Platform

Web + Admin + API Layer

Approach

AI-Assisted Development

Project Overview

A full-stack AI product ecosystem, not just an app.

Gardenly is a digital marketplace platform designed to connect customers with gardening services, vendors, and products within the UAE. Built as a scalable ecosystem covering customer experience, vendor onboarding, admin operations, and backend API architecture.

The project was executed using an AI-assisted product development approach, where planning, structuring, and execution were accelerated through intelligent tooling while maintaining strong product thinking and system design.

πŸ›’

Customer Layer

Discovery, booking & purchase

πŸͺ

Vendor Layer

Onboarding, listing & fulfillment

βš™οΈ

Admin Layer

Control, approvals & operations

πŸ”Œ

API Layer

Data consistency & scalability

Problem / Opportunity

The Gaps in the Market

  • Fragmented service discovery for gardening solutions
  • No structured vendor onboarding and approval workflow
  • Lack of consistency between admin, customer, and backend data flows
  • No scalable architecture to support marketplace growth
  • Poor visibility into service lifecycle, booking, and pricing logic
My Role

AI Product Manager &
Full Stack Development Lead

I owned the product from concept to structured execution β€” spanning strategy, system design, and development.

  • Defining product vision and scope
  • Designing complete user and system workflows
  • Structuring backend and API architecture
  • Driving AI-assisted development execution
  • Aligning frontend, backend, and admin systems
  • Designing vendor onboarding and pricing models
  • Supporting product marketing and growth direction
Product Strategy30%
System Architecture25%
Development Execution30%
Product Marketing15%
Product Thinking

Multi-Actor Ecosystem Design

Instead of building a simple booking platform, I approached Gardenly as a multi-actor ecosystem with clear separation of concerns at each layer.

System Architecture Overview

FRONTENDBACKENDDATABASECustomer WebNext.jsAdmin PortalNext.jsFastAPI CoreAPI GatewayAuth ServiceJWT + SessionsVendor ServiceOnboardingBooking LogicOrders + PaymentPostgreSQL DatabaseRelational Single Source of Truth
Key Features Delivered

Full-Spectrum Feature Coverage

  • Service discovery and intelligent search
  • Service booking with dynamic pricing
  • Product catalog browsing and purchase
  • Lead capture and enquiry system
  • Real-time pricing visibility
Core Workflow

End-to-End System Flow

How data and operations move across all four actor layers β€” from a customer browsing to a vendor completing a service.

End-to-End Product Workflow

CUSTOMERVENDORADMINBACKENDBrowseServices01BookService02CreateBooking03MonitorBooking04ReceiveAssignment05CompleteService06CloseBooking07
AI Workflow

AI as a Force Multiplier

AI was embedded across the entire product lifecycle β€” not just for coding but as a strategic execution tool.

πŸ“‹

Requirement Structuring

Breaking down ambiguous goals into clear, executable specs.

πŸ—ΊοΈ

Workflow & System Design

Mapping user journeys and system interactions intelligently.

πŸ”Œ

API Planning

Designing endpoint contracts and validating data schemas.

✨

UX Refinement

Rapidly prototyping and iterating on interaction patterns.

⚑

Dev Acceleration

Generating boilerplate, tests, and logic with AI assistance.

πŸ”

Debug & Optimize

Identifying bottlenecks and improving system reliability.

Data Architecture

Data Model & Schema Thinking

A clean relational schema built to maintain data integrity across all actors and service types.

Data Model & Schema Architecture

Useridemailroleaddress_idVendoriduser_idtypestatusServiceidvendor_idtitlepriceBookingiduser_idservice_idstatusOrderiduser_idproduct_idtotalPaymentidorder_idamountstatusProductidvendor_idnamepriceAddressiduser_idcityarea
Challenges & Decisions

Key Architectural Decisions

01

Maintaining consistency between admin-created and user-submitted data

02

Designing a scalable vendor onboarding model that handles both individuals and companies

03

Structuring API-first architecture for future mobile expansion

04

Balancing execution speed (AI-driven) with system clarity and documentation

05

Avoiding hardcoded flows and ensuring dynamic, configurable scalability

Outcomes / Impact

What Was Delivered

πŸ—οΈ

Product Foundation

Built a structured, scalable architecture from a blank canvas.

πŸ”„

Clear Actor Workflows

Defined flows across Customer, Vendor, Admin, and API layers.

πŸ“

Scalable Architecture

Marketplace design ready for growth and future modules.

πŸ€–

AI-Led Execution

Accelerated the full development cycle with AI tooling.

πŸ“‘

API-First System

Ready for mobile and third-party integrations from day one.

πŸ’Ό

Investment-Ready

System structured and documented for stakeholder clarity.

Final Reflection

Strategy + Architecture + AI Execution

This project represents my ability to operate as an AI Product Manager across strategy, system design, and execution β€” combining product thinking, technical architecture, AI-assisted development, and business understanding to deliver scalable digital products.

Product Thinking
Technical Architecture
AI-Assisted Development
Business Understanding