My Role
AI product and data-integration engineer
All Projects
Interactive prototypeFlagship case study
Travel Intelligence
AI Travel Platform
Travel search ranks options without clearly exposing how fare, time, comfort, and traveler preferences influenced the recommendation.
Interactive preview
Full demoWhat I Built
AI-powered travel intelligence platform that combines flight, destination, transportation, and trip-planning information into a unified experience. Designed to help users research destinations, evaluate travel options, and turn fragmented travel data into personalized recommendations and actionable itineraries.
Evidence
Working interface, documented system behavior, and implementation-level decisions.
Technical Architecture
From system input to explainable output.
The control gate is shown as a first-class stage, not an afterthought added around the workflow.
Technical Decisions
- Built the comfort index from stop count, red-eye status, and seat pitch rather than treating price and duration as the only variables.
- Implemented weighted preference profiles for value, cheapest, fastest, and most comfortable, each with its own blend of the three axes.
Security Considerations
- The public case study makes no additional security claim beyond the documented controls shown in the architecture.
Outcome & Evidence
- Normalised fare, elapsed time, and a derived comfort index across the candidate set so three incompatible axes can be ranked together.
AITravel TechnologyAPIsData IntegrationIntelligent Search
Working product
Try the interactive demo.
The product experience is part of this case study. Explore it here, reset its state, or switch viewport sizes without leaving the project page.
travel-intelligence.zainkhalilkhan.com
Travel Intelligence
AI product
Local inference mock
Travel IntelligenceWorkspace3 updates
Travel Intelligence · Itinerary RankingORD → DXB · 6 itineraries under the best value preference
66% match
Six live-shaped itineraries for one route. Fare, elapsed time, and a comfort index built from stop count, red-eye status, and seat pitch are normalised across the candidate set, then blended by the active traveller preference so a single score can rank all three axes at once.
Balances fare against total travel time. Weights: fare 50%, time 30%, comfort 20%.
Recommended
Delta
DL-8840
Fare
$338
cheapest is $264
Elapsed
10h 40m
fastest is 8h 20m
Match
66%
against this preference
Best matchDL-8840
66%match
Match score per itineraryfare 50% · time 30% · comfort 20%
Reordering these bars is the whole product: the same six itineraries rank differently under each preference, and the cheapest is rarely the best match.
Delta via ATL wins on the blend, trading 95 minutes on the ground for $74 over the cheapest fare.
Itineraries
Zain Khalil Khan
Client-side sandbox. State is in memory and nothing is sent to a server.
Next Case Study