Case study
Gledva — AI Travel Companion
An intelligent travel workspace that keeps trip context and assists the traveler throughout the journey.
- Role
- Founder & Product Manager
- Product
- B2C AI travel platform
- Platforms
- iOS · Android · Web
- Stage
- 0→1 → Launch → Iteration
The challenge
Travelers manage trips across fragmented systems — booking confirmations, itineraries, documents, calendars, packing lists, budgets, maps and travel applications. I identified an opportunity to build an intelligent travel workspace that maintains trip context and assists the traveler throughout the journey.
What I owned
Product strategy → user experience → requirements → AI architecture → engineering prioritization → launch → GTM → analytics → iteration.
Capabilities I defined and prioritized
From AI chat to agentic travel
Moving beyond a standalone chatbot toward an architecture that understands context and supports multi-step workflows.
- User intent
- Context + trip state
- LLM reasoning
- Orchestration
- RAG / retrieval
- Tools + APIs
- Action
- Memory / state
- Guardrails + human approval
- Evaluation + observability
Traveler: “It looks like rain tomorrow. Reorganize my day.”
- 01
Understand
Identify destination, date, user intent and the requested change.
- 02
Retrieve
Pull tomorrow's itinerary, bookings, activity details and relevant traveler preferences.
- 03
Use tools
Obtain relevant external information such as weather and activity data through appropriate services.
- 04
Reason
Determine which activities are weather-sensitive and identify alternatives.
- 05
Generate plan
Create a revised itinerary considering timing, location and existing commitments.
- 06
Ask for approval
Present consequential itinerary changes before modifying confirmed plans.
- 07
Execute
Update approved itinerary elements, tasks or reminders through deterministic application functions.
- 08
Maintain state
Persist the updated trip so subsequent AI interactions use the latest information.
The product decision
Not the question
“How do I add AI to a travel app?”
The real question
“Which decisions should AI make, which actions should deterministic software execute, and when should the traveler remain in control?”
Designing AI autonomy
I designed the experience around levels of risk.
Recommend nearby restaurants.
AI recommends autonomously.
Reorganize tomorrow's itinerary.
Generate proposed changes → request confirmation → apply changes.
Cancel a hotel or purchase a new booking.
Never execute from inferred intent → require explicit authorization.
Product principle: the higher the consequence and the lower the reversibility, the stronger the human approval requirement.
AI product architecture
Shipped capabilities and designed architecture, labelled honestly.
Reasoning
shippedLLMs interpret natural-language requests and reason across ambiguous travel requirements.
Retrieval
designedTraveler-specific information is retrieved from structured trip data and documents instead of relying on model knowledge.
Tool calling
designedExternal and deterministic capabilities are exposed as controlled tools for the AI workflow.
State
shippedStructured information for destinations, dates, activities, bookings, documents, tasks, expenses, packing and reminders.
Guardrails
designedHigh-consequence actions require stronger controls and explicit user approval.
Failure recovery
designedWhen reliable information cannot be retrieved, the product fails safely instead of allowing the model to fabricate an answer.
AI evaluation framework
I evaluate AI experiences across the complete workflow rather than only the generated text.
| Dimension | Product question |
|---|---|
| Task success | Did the AI accomplish the user's objective? |
| Groundedness | Is the response supported by trusted information? |
| Tool accuracy | Did the agent select and call the correct tool? |
| Safety | Did it stay within permitted actions? |
| Latency | Was the experience responsive enough? |
| Cost | What did successful task completion cost? |
| User satisfaction | Did the customer find the result valuable? |
Optimization objective: Quality × Reliability × Latency × Cost
The goal isn't to use the largest or most expensive model. The goal is to select the architecture that meets the required product experience.
Results
Key learning
Building an AI product isn't simply adding a chatbot. The deeper product challenge is determining where probabilistic reasoning creates value, where deterministic software should remain in control, how the system should access trusted information, and how much autonomy customers should give AI.