View Proposal

Proposer
Zi Hau Chin
Title
From Hand-Drawn Sketches to Business Applications: Evaluating Agentic AI-Assisted Frontend Development on the Frappe Platform
Goal
To investigate whether a functional and usable Frappe-based business application frontend can be developed from hand-drawn interface sketches, defined business workflows and API specifications using Agentic AI, with limited manual programming. The project will also identify the business, technical and analytical knowledge required to supervise and validate the AI-assisted development process.
Description
The student will study an existing Frappe-based webshop backend and develop a compatible frontend using Agentic AI tools. The student will be provided with, or develop: -A business scenario -Required business processes -Functional and non-functional requirements -Hand-drawn interface sketches -User navigation flows -Relevant Frappe DocTypes -API specifications The student will use these artefacts as inputs to an AI-assisted development process. The AI agent may generate frontend code, API calls, page structures and interface components. The student must understand, test and validate all generated outputs. The investigation must document: -What information the AI agent required. -How sketches were interpreted. -How requirements were communicated. -How APIs were identified and integrated. -What errors or unsupported assumptions were generated. -How much human intervention was required. -Whether the final application satisfied the requirements. -Whether the resulting process is repeatable. The student should propose a reusable process such as: Business Analysis → Requirements → Hand-Drawn Sketches → API Mapping → AI Instructions → Generation → Human Validation → Testing → Refinement
Resources
Frappe, FrappeWebshop; Business-process modelling resources;
Background
AI coding agents are changing application development by allowing users to describe requirements, interfaces and behaviours using natural language. However, successful business application development still requires an understanding of business processes, system requirements, data structures, APIs, user interaction and validation. This is relevant to HWUM’s Information Systems with Data Analytics programme, which covers information systems, data management, business processes, digital transformation, programming, problem-solving and professional practice. Frappe provides a low-code framework built around DocTypes and includes administrative interfaces, permissions, workflows and API capabilities. Its documentation states that REST APIs are generated for DocTypes and support standard data operations. This project investigates whether an Information Systems student can act as an AI-assisted application developer by defining what the system should do, understanding its backend and APIs, directing AI agents, and critically validating the resulting application.
Url
Difficulty Level
High
Ethical Approval
None
Number Of Students
1
Supervisor
Zi Hau Chin
Keywords
ai-assisted development, frapep framework, low-code development, business applications, human-ai collaboration, digital transformation
Degrees
Back to List