Honda - Koo
Honda - Koo
Imagine walking into a massive trade show and instantly knowing where to go, who to meet, and what sessions to attend—without lifting a finger. Koo transforms chaos into clarity with AI-powered personalization, navigation, and note-taking.
VIsion & Problem Statement
VIsion & Problem Statement
Vision:
To redefine the trade show experience through AI, empowering attendees to navigate, connect, and absorb information effortlessly while giving organizers a powerful tool to optimize operations and engagement.
Problem Statement:
Trade shows are overwhelming: attendees often face disorganized schedules, confusing layouts, missed networking opportunities, and fragmented information capture. Traditional solutions lack personalization, adaptability, and real-time support. Koo solves this by combining LLMs, LVMs, and multimodal AI into one smart assistant.
Vision:
To redefine the trade show experience through AI, empowering attendees to navigate, connect, and absorb information effortlessly while giving organizers a powerful tool to optimize operations and engagement.
Problem Statement:
Trade shows are overwhelming: attendees often face disorganized schedules, confusing layouts, missed networking opportunities, and fragmented information capture. Traditional solutions lack personalization, adaptability, and real-time support. Koo solves this by combining LLMs, LVMs, and multimodal AI into one smart assistant.
VIsion & Problem Statement
Vision:
To redefine the trade show experience through AI, empowering attendees to navigate, connect, and absorb information effortlessly while giving organizers a powerful tool to optimize operations and engagement.
Problem Statement:
Trade shows are overwhelming: attendees often face disorganized schedules, confusing layouts, missed networking opportunities, and fragmented information capture. Traditional solutions lack personalization, adaptability, and real-time support. Koo solves this by combining LLMs, LVMs, and multimodal AI into one smart assistant.
Product Goal
Product Goal
Develop an intelligent, AI-driven attendee assistant that generates personalized itineraries, enables real-time navigation, synthesizes notes, and recommends networking connections—revolutionizing the trade show experience for both attendees and organizers.
Develop an intelligent, AI-driven attendee assistant that generates personalized itineraries, enables real-time navigation, synthesizes notes, and recommends networking connections—revolutionizing the trade show experience for both attendees and organizers.
Product Goal
Develop an intelligent, AI-driven attendee assistant that generates personalized itineraries, enables real-time navigation, synthesizes notes, and recommends networking connections—revolutionizing the trade show experience for both attendees and organizers.
User Stories
User Stories
Title | As a/an | I want to | So that |
|---|---|---|---|
Personalized Itinerary | Attendee | Receive a tailored event itinerary based on my preferences | I can focus on relevant sessions without manually planning |
Real-Time Navigation | Attendee | Use a dynamic indoor map to find my way | I don’t get lost or miss key sessions |
Smart Note Taking | Attendee | Capture, synthesize, and organize information across media types | I can save time and retain important insights without manual effort |
Intelligent Networking | Attendee | Discover and connect with relevant attendees | I can build meaningful relationships that align with my goals |
Event Analytics | Organizer | View attendee behaviors and engagement metrics | I can optimize future event planning and operations |
In-App Promotions | Exhibitor | Advertise directly to interested attendees through the app | I can increase visibility and conversions effectively |
Title | As a/an | I want to | So that |
|---|---|---|---|
Personalized Itinerary | Attendee | Receive a tailored event itinerary based on my preferences | I can focus on relevant sessions without manually planning |
Real-Time Navigation | Attendee | Use a dynamic indoor map to find my way | I don’t get lost or miss key sessions |
Smart Note Taking | Attendee | Capture, synthesize, and organize information across media types | I can save time and retain important insights without manual effort |
Intelligent Networking | Attendee | Discover and connect with relevant attendees | I can build meaningful relationships that align with my goals |
Event Analytics | Organizer | View attendee behaviors and engagement metrics | I can optimize future event planning and operations |
In-App Promotions | Exhibitor | Advertise directly to interested attendees through the app | I can increase visibility and conversions effectively |
User Stories
Title | As a/an | I want to | So that |
|---|---|---|---|
Personalized Itinerary | Attendee | Receive a tailored event itinerary based on my preferences | I can focus on relevant sessions without manually planning |
Real-Time Navigation | Attendee | Use a dynamic indoor map to find my way | I don’t get lost or miss key sessions |
Smart Note Taking | Attendee | Capture, synthesize, and organize information across media types | I can save time and retain important insights without manual effort |
Intelligent Networking | Attendee | Discover and connect with relevant attendees | I can build meaningful relationships that align with my goals |
Event Analytics | Organizer | View attendee behaviors and engagement metrics | I can optimize future event planning and operations |
In-App Promotions | Exhibitor | Advertise directly to interested attendees through the app | I can increase visibility and conversions effectively |
Core Features
Core Features
Feature | Description | Priority |
|---|---|---|
Personalized Itinerary | LLM-generated and updated based on preferences, time, and session dynamics | P1 |
Indoor Navigation | Real-time, AI-enhanced navigation through multi-floor, crowded venues | P1 |
Multimodal Note Capture | Enables note-taking using text, images, and voice, with auto-synthesis | P1 |
Smart Networking | Suggests contacts based on interest, role, or company—includes QR code linking | P2 |
AI-Powered Notifications | Alerts for upcoming sessions, hydration reminders, and location-based prompts | P2 |
Feature | Description | Priority |
|---|---|---|
Personalized Itinerary | LLM-generated and updated based on preferences, time, and session dynamics | P1 |
Indoor Navigation | Real-time, AI-enhanced navigation through multi-floor, crowded venues | P1 |
Multimodal Note Capture | Enables note-taking using text, images, and voice, with auto-synthesis | P1 |
Smart Networking | Suggests contacts based on interest, role, or company—includes QR code linking | P2 |
AI-Powered Notifications | Alerts for upcoming sessions, hydration reminders, and location-based prompts | P2 |
Core Features
Feature | Description | Priority |
|---|---|---|
Personalized Itinerary | LLM-generated and updated based on preferences, time, and session dynamics | P1 |
Indoor Navigation | Real-time, AI-enhanced navigation through multi-floor, crowded venues | P1 |
Multimodal Note Capture | Enables note-taking using text, images, and voice, with auto-synthesis | P1 |
Smart Networking | Suggests contacts based on interest, role, or company—includes QR code linking | P2 |
AI-Powered Notifications | Alerts for upcoming sessions, hydration reminders, and location-based prompts | P2 |
Success Metrics
Success Metrics
Metric | Description |
|---|---|
Itinerary Match Score | % of attendees satisfied with suggested schedule |
Navigation Accuracy | % of users reaching locations without assistance |
Note Conversion Rate | % of saved notes turned into summarized or exported format |
Connection Conversion | % of suggested connections that result in LinkedIn exchanges or saved contact |
Session Attendance Lift | Increase in average sessions attended compared to manual planning |
Organizer Satisfaction Score | Feedback on how Koo streamlined operations and insights |
Metric | Description |
|---|---|
Itinerary Match Score | % of attendees satisfied with suggested schedule |
Navigation Accuracy | % of users reaching locations without assistance |
Note Conversion Rate | % of saved notes turned into summarized or exported format |
Connection Conversion | % of suggested connections that result in LinkedIn exchanges or saved contact |
Session Attendance Lift | Increase in average sessions attended compared to manual planning |
Organizer Satisfaction Score | Feedback on how Koo streamlined operations and insights |
Success Metrics
Metric | Description |
|---|---|
Itinerary Match Score | % of attendees satisfied with suggested schedule |
Navigation Accuracy | % of users reaching locations without assistance |
Note Conversion Rate | % of saved notes turned into summarized or exported format |
Connection Conversion | % of suggested connections that result in LinkedIn exchanges or saved contact |
Session Attendance Lift | Increase in average sessions attended compared to manual planning |
Organizer Satisfaction Score | Feedback on how Koo streamlined operations and insights |
Technical Stack
Technical Stack
Models:
LLM: Personalized Itinerary Generation
LVM: Floor Plan Recognition & Mapping
Multimodal AI: Note Synthesis (text, voice, image)
Frameworks & Tools:
Cloud-based AI Infrastructure
Indoor Mapping APIs
QR-Based Networking Tools
Outputs:
Custom Itineraries
Synthesized Notes & Event Summaries
Real-Time Maps & Alerts
Attendee Behavior Analytics
Models:
LLM: Personalized Itinerary Generation
LVM: Floor Plan Recognition & Mapping
Multimodal AI: Note Synthesis (text, voice, image)
Frameworks & Tools:
Cloud-based AI Infrastructure
Indoor Mapping APIs
QR-Based Networking Tools
Outputs:
Custom Itineraries
Synthesized Notes & Event Summaries
Real-Time Maps & Alerts
Attendee Behavior Analytics
Technical Stack
Models:
LLM: Personalized Itinerary Generation
LVM: Floor Plan Recognition & Mapping
Multimodal AI: Note Synthesis (text, voice, image)
Frameworks & Tools:
Cloud-based AI Infrastructure
Indoor Mapping APIs
QR-Based Networking Tools
Outputs:
Custom Itineraries
Synthesized Notes & Event Summaries
Real-Time Maps & Alerts
Attendee Behavior Analytics
Key Results
90 %+ itinerary satisfaction from pilot testers
85% of users found indoor navigation highly helpful
100% desired QR-based contact exchange over traditional methods
75% emphasized the need for post-event summaries and follow-ups
Achieved real-time navigation accuracy of 95% in a 3-floor test event
Constraints, Risks, and Mitigations
Issue / Constraint | Type | Mitigation / Notes |
|---|---|---|
Floor plans vary drastically in format | Constraint | Use LVM to interpret diverse visual inputs, reducing manual mapping effort |
LLM output accuracy for itinerary curation | Risk | Use human-curated training prompts and continuous model fine-tuning |
User adoption in conservative B2B environments | Risk | Leverage organizer partnerships for trust-building and forced onboarding |
Data privacy & security for attendee info | Constraint | Anonymize all personal data and adhere to GDPR/CCPA standards |
Venue connectivity may impact navigation | Risk | Preload maps and use hybrid GPS + proximity sensor fallback |
Business Impact
Boosts attendee satisfaction and retention across multi-day events
Enhances the organizer's value proposition through premium features
Opens up targeted monetization through exhibitor promotions
Provides organizers with deep insights for optimizing future events
Differentiates venues and events with cutting-edge tech adoption
Future Roadmap
Short-Term
MVP rollout with 3 pilot trade shows
Refine LLM and navigation models based on feedback
Add export options for notes and itineraries
Mid-Term
Integrate with CRM and ticketing platforms
Launch live heatmaps for crowd navigation
Expand networking to include interest-based speed meeting
Long-Term
Enable voice-based AI concierge for on-site help
Offer exhibitor booth insights through heat mapping
Expand to international expos with multilingual support
Key Results
90 %+ itinerary satisfaction from pilot testers
85% of users found indoor navigation highly helpful
100% desired QR-based contact exchange over traditional methods
75% emphasized the need for post-event summaries and follow-ups
Achieved real-time navigation accuracy of 95% in a 3-floor test event
Constraints, Risks, and Mitigations
Issue / Constraint | Type | Mitigation / Notes |
|---|---|---|
Floor plans vary drastically in format | Constraint | Use LVM to interpret diverse visual inputs, reducing manual mapping effort |
LLM output accuracy for itinerary curation | Risk | Use human-curated training prompts and continuous model fine-tuning |
User adoption in conservative B2B environments | Risk | Leverage organizer partnerships for trust-building and forced onboarding |
Data privacy & security for attendee info | Constraint | Anonymize all personal data and adhere to GDPR/CCPA standards |
Venue connectivity may impact navigation | Risk | Preload maps and use hybrid GPS + proximity sensor fallback |
Business Impact
Boosts attendee satisfaction and retention across multi-day events
Enhances the organizer's value proposition through premium features
Opens up targeted monetization through exhibitor promotions
Provides organizers with deep insights for optimizing future events
Differentiates venues and events with cutting-edge tech adoption
Future Roadmap
Short-Term
MVP rollout with 3 pilot trade shows
Refine LLM and navigation models based on feedback
Add export options for notes and itineraries
Mid-Term
Integrate with CRM and ticketing platforms
Launch live heatmaps for crowd navigation
Expand networking to include interest-based speed meeting
Long-Term
Enable voice-based AI concierge for on-site help
Offer exhibitor booth insights through heat mapping
Expand to international expos with multilingual support
Key Results
90 %+ itinerary satisfaction from pilot testers
85% of users found indoor navigation highly helpful
100% desired QR-based contact exchange over traditional methods
75% emphasized the need for post-event summaries and follow-ups
Achieved real-time navigation accuracy of 95% in a 3-floor test event
Constraints, Risks, and Mitigations
Issue / Constraint | Type | Mitigation / Notes |
|---|---|---|
Floor plans vary drastically in format | Constraint | Use LVM to interpret diverse visual inputs, reducing manual mapping effort |
LLM output accuracy for itinerary curation | Risk | Use human-curated training prompts and continuous model fine-tuning |
User adoption in conservative B2B environments | Risk | Leverage organizer partnerships for trust-building and forced onboarding |
Data privacy & security for attendee info | Constraint | Anonymize all personal data and adhere to GDPR/CCPA standards |
Venue connectivity may impact navigation | Risk | Preload maps and use hybrid GPS + proximity sensor fallback |
Business Impact
Boosts attendee satisfaction and retention across multi-day events
Enhances the organizer's value proposition through premium features
Opens up targeted monetization through exhibitor promotions
Provides organizers with deep insights for optimizing future events
Differentiates venues and events with cutting-edge tech adoption
Future Roadmap
Short-Term
MVP rollout with 3 pilot trade shows
Refine LLM and navigation models based on feedback
Add export options for notes and itineraries
Mid-Term
Integrate with CRM and ticketing platforms
Launch live heatmaps for crowd navigation
Expand networking to include interest-based speed meeting
Long-Term
Enable voice-based AI concierge for on-site help
Offer exhibitor booth insights through heat mapping
Expand to international expos with multilingual support
More Works More Works
More Works More Works
OPEN AI - RED TEAMING
RED TEAMING + GEN AI + CHATGPT
2024
2024
OPEN AI - RED TEAMING
RED TEAMING + GEN AI + CHATGPT
2024
2024
OPEN AI - RED TEAMING
RED TEAMING + GEN AI + CHATGPT
2024
2024
OPEN AI - RED TEAMING
RED TEAMING + GEN AI + CHATGPT
2024
2024
CMU - GEN AI LAB RAG
RAG SYSTEM + SEARCH + DB
2024
2024
CMU - GEN AI LAB RAG
RAG SYSTEM + SEARCH + DB
2024
2024
CMU - GEN AI LAB RAG
RAG SYSTEM + SEARCH + DB
2024
2024
CMU - GEN AI LAB RAG
RAG SYSTEM + SEARCH + DB
2024
2024