Back to Techathon 6.0

Techathon 6.0

Problem Statement 5: Retail

About the business:

A leading retail brand seeks to revolutionize its sales experience by deploying an AI-driven Conversational Sales Agent that seamlessly operates across online and physical channels. This Agent must emulate a top-tier sales associate—guided by natural, personalized dialogue—while orchestrating specialized Worker Agents to handle tasks from product discovery through checkout and post-purchase support.

Problem statement:

Customers face fragmented experiences when moving between online browsing, mobile app shopping, messaging apps, and in-store interactions. Limited bandwidth among sales associates leads to missed up-sell and cross-sell opportunities. The goal is to increase Average Order Value (AOV) and conversion rates by offering a unified, human-like conversational journey that anticipates needs, provides tailored recommendations and facilitates sales across all channels.

Goal:

The team should design an Agentic AI solution where the Sales Agent:

  • Engages customers via web chat, mobile app, WhatsApp/Telegram, in-store kiosk or voice assistant
  • Understands preferences and context (e.g., past purchases, store location, current promotions) and guides customers toward relevant products
  • Coordinates multiple Worker Agents to handle inventory checks, personalized recommendations, promotions lookup, payment processing and order confirmation—culminating in a seamless purchase or booking (e.g., “reserve in-store for try-on”)

Key deliverable:

A live demo or recorded walkthrough (maximum four minutes) showcasing:

  • A customer initiating a conversation on one channel (e.g., mobile app)
  • The Sales Agent adapting the conversation when the customer switches to an in-store kiosk or messaging app
  • End-to-end orchestration: recommendation → inventory check → payment → digital/physical fulfillment → post-purchase follow-up

Agentic AI Roles
Role
Responsibilities

Sales Agent

Manages multi-channel conversation flow
Routes tasks to Worker Agents
Handles context switching and session continuity

Worker Agents

Recommendation Agent

Analyzes customer profile, browsing history and seasonal trends
Suggests products, bundles and promotions

Inventory Agent

Checks real-time stock across warehouses and stores
Offers “ship to home,” “click & collect,” or “in-store availability”

Payment Agent

Processes payments via saved cards, UPI, gift cards or in-store POS
Manages payment failures and retries

Fulfillment Agent

Schedules delivery or reserve-in-store slots
Notifies logistics or store staff for pickup orders

Loyalty and Offers Agent

Applies loyalty points, coupon codes and personalized offers
Calculates final pricing and displays savings

Post-Purchase Support Agent

Handles returns/exchanges, tracks shipments and solicits feedback

Data and system assumptions

  • Synthetic customer profiles: ≥ 10 customers with demographics, purchase history, loyalty tier, device preferences
  • Product catalog API: Mock endpoint with SKUs, categories, attributes, pricing, images
  • Inventory server: Simulated real-time stock levels for online and multiple store locations
  • Payment gateway stub: Dummy API for authorizations, captures and declined transactions
  • Loyalty and promotions service: Mock rules engine for loyalty points and timed promotions
  • POS integration: Simulated in-store terminal interactions for barcode scan and payment

Evaluation Criteria
Criterion
Weightage in %

Technical design

35%

Clear orchestration using an Agentic AI framework (LangGraph, AutoGen, etc.)

Robust channel-handoff logic

Personalization and UX

25%

Natural, context-aware dialogues
Seamless channel transitions

Data and workflow realism

20%

Quality of synthetic datasets and APIs
Real-world–like inventory and payment flows

Demo and storytelling

20%

Engaging, end-to-end scenario demonstrating key features
Handling of edge cases (out-of-stock, payment decline, returns)

Submission Format

  • Demo:
    • Live prototype or three to four minute video walkthrough
  • Documentation
    • System architecture diagram (Mermaid or equivalent)
    • Agent roles and workflows
    • Data schema and API assumptions (sample JSON/CSV)
    • Channel-handoff logic description

Tips for participants

  • Omnichannel consistency: Maintain session continuity when moving between channels (e.g., chat → in-store kiosk)
  • Sales psychology: Use persuasive, consultative language—ask open questions (“What occasion are you shopping for?”), suggest complementary items (“These shoes pair well with…”) and handle objections gracefully
  • Edge-case demonstrations: Show how the Agent recovers from scenarios like payment failures, out-of-stock items or a request to modify an order
  • Modular orchestration: Keep Worker Agents loosely coupled so new capabilities (e.g., Gift-wrapping Agent) can be added easily

Previous    

  Problem Statement 4          

Next

  EY Techathon 6

Contact us
Like what you’ve seen? Get in touch to learn more.

EY refers to the global organization, and may refer to one or more, of the member firms of Ernst & Young Global Limited, each of which is a separate legal entity. Ernst & Young Global Limited, a UK company limited by guarantee, does not provide services to clients.