RetailFlow
Overview
Section titled “Overview”RetailFlow is a simulated Australian multi-channel retailer — selling fashion, homewares, and lifestyle products across physical stores and online. Students interact with AI chatbot employees to practise professional skills in a retail business context.
Website: github.com/michael-borck/retailflow
The Organisation
Section titled “The Organisation”RetailFlow is a mid-sized Australian retailer operating both brick-and-mortar stores and an online shop. Like any real retail business, it has:
- A leadership team managing growth and strategy
- Operations staff handling supply chain and logistics
- Customer service representatives dealing with real issues
- Internal processes, policies, and organisational challenges
AI Employees
Section titled “AI Employees”Each RetailFlow employee is an AI chatbot with a unique backstory, personality, and role. Students interview leadership and staff to gather information, just as they would in a real consulting engagement.
The information students need is distributed across multiple people — understanding the full picture requires talking to several employees and reconciling different perspectives.
Use Cases
Section titled “Use Cases”Retail Operations Analysis
Section titled “Retail Operations Analysis”Interview staff across the business to understand how RetailFlow operates. Map processes from warehouse to customer delivery. Identify bottlenecks and recommend improvements.
Marketing & Customer Research
Section titled “Marketing & Customer Research”Talk to customer-facing staff and leadership to understand RetailFlow’s market position, customer segments, and marketing strategy. Develop evidence-based recommendations.
Supply Chain & Logistics
Section titled “Supply Chain & Logistics”Explore how RetailFlow manages inventory, suppliers, and distribution across multiple channels. Identify risks and inefficiencies through staff interviews.
Business Analysis
Section titled “Business Analysis”Conduct a broad organisational analysis. Understand the challenges RetailFlow faces, gather requirements for new systems or processes, and develop a business case.
Technology
Section titled “Technology”RetailFlow is powered by:
- Fine-tuned small LLMs — each character has been fine-tuned to stay in role
- Ollama — local inference, no cloud dependency
- Character-specific system prompts — backstory, personality traits, knowledge boundaries
- Conversation memory — characters remember what you’ve discussed in the current session
All inference runs locally. No student conversations are sent to external APIs.
Getting Started
Section titled “Getting Started”Visit the RetailFlow site to start interacting with RetailFlow employees. Your educator will provide specific instructions for your unit’s assessment tasks.