Transport Plan

Below this text 2 projects i worked on at Transport Plan. Automated UI testing & an AI chatbot.

Automated UI Testing in WinForms.NET

In this project, I implemented automated UI testing for a WinForms.NET application using Reqnroll and FlaUI. The goal was to create a prototype to improve software reliability and make testing accessible for both developers and non-technical coworkers.

Part 1 – Reqnroll (BDD Testing)

Reqnroll is a Behavior-Driven Development (BDD) framework that allows tests to be written in a human-readable format using Gherkin syntax (Given, When, Then).

This makes tests easy to understand and write, even for people without a programming background.

Why this matters

By using Reqnroll, I made it possible for coworkers to:

Understand what the system should do
Help write and review test scenarios
Validate business rules
Use tests as living documentation

This improves communication and ensures the software matches real business needs.

Part 2 – FlaUI (WinForms UI Automation)

FlaUI is a Windows UI automation library that allows tests to interact with real interface elements like buttons, forms, and tables.

Why FlaUI is important

While Reqnroll defines what should happen, FlaUI executes those actions in the real application.

It allows me to:

Simulate real user behavior
Test full application flows
Catch UI bugs early
Prevent regressions

This results in more reliable releases and less manual testing.

Part 3 – Working with WinForms + DevExpress

This application used a combination of standard WinForms components and DevExpress UI controls.

This made UI automation more complex because:

DevExpress elements behave differently from native WinForms controls
UI structures were more nested and dynamic
Some controls required custom handling

I adapted my FlaUI logic to correctly recognize and interact with these mixed UI elements, making the automation system more robust and flexible.

Part 4 – Azure DevOps Workflow

For version control and collaboration, I worked with Azure DevOps.

I:

Created my own feature branches
Developed new functionality independently
Opened pull requests
Handled feedback and merges

This ensured that my changes were reviewed, traceable, and safely integrated into the main codebase.

Summary

By combining Reqnroll, FlaUI, and a structured Azure DevOps workflow, I built a testing system that is:

Reliable
Scalable
Understandable for non-programmers
Suitable for complex WinForms + DevExpress interfaces

This approach improved software quality and reduced manual testing effort.

Internal Knowledge Chatbot (n8n)

I built a custom internal chatbot using n8n that centralizes all available knowledge about the company’s software into a single interface. Instead of searching through multiple systems manually, users can now ask questions in natural language and receive relevant answers instantly.

The chatbot integrates multiple data sources, including:

Zendesk tickets (open and closed)
Zendesk help articles
DevOps wiki pages
Internal documentation

All this information is processed, cleaned, and stored in a vector database, allowing the chatbot to retrieve relevant context before generating a response.
Internal Zendesk JavaScript code snippet

Why this chatbot is different

This is not a simple Q&A bot. It is a fully automated workflow system that:

Authenticates users through a login system
Retrieves live data from multiple platforms
Converts documentation into searchable embeddings
Uses semantic search to find relevant content
Generates contextual AI responses
Returns formatted HTML output

This makes it a practical internal tool rather than a demo project.

Why this is valuable for a company

This chatbot significantly reduces time spent searching for information and asking colleagues repetitive questions. It improves onboarding, support, and knowledge sharing by making internal documentation easily accessible.

It also ensures that knowledge does not get lost across separate platforms, but instead becomes part of one intelligent system.