INF Weather
INF Weather is an exploratory AI dashboard designed for weather and air traffic planning. It helps operators make clearer decisions by combining AI-based weather forecasts with a visual display of flight routes and atmospheric data. This demo showcases the potential of the Fuxi weather model in aviation scenarios.
Challenge & Task
We partnered with Spring Airlines to explore how AI weather models could support route planning. In practice, their dispatchers face several challenges:
Fragmented/Inaccurate Data: Weather data is often fragmented and hard to interpret.
Separated platforms: Disconnected weather and aviation data sources require constant toggling between systems.
Complex System: Planners struggle to connect weather patterns with route disruptions or delays.
This project aimed to make those relationships more and visible. Working with product managers and airline stakeholders, we defined key user needs.
Combine weather info: Show multiple weather patterns in one place.
Filter clearly: Better control over what weather data is shown.
Connect information: Help users connect weather patterns to delays or route changes.
Design
After researching how weather is typically presented in aviation tools (e.g. ForeFlight, FAA GFA), I designed a map-based interface that simplifies complex, layered information and makes weather impact on flights more visible.
I focused on these areas:
Easy switches on weather data
Weather risks, such as cloud coverage, thunderstorms or wind can be easily filtered on map. Users can switch between weather layers with a single click, and see changes reflected instantly on the map.
Multi-layer map and interactive timeline
I designed the interface to support layered filtering, with controls that let users isolate specific information like route, route alerts or airports. Below the map, a synced timeline shows how on-time rates and weather patterns evolve over time, helping planners anticipate issues.
Route-focused weather filtering
Users can select any route to see weather patterns along its path, and then filter by specific weather conditions (e.g., low clouds only) to evaluate risk. This lets them clearly see how each weather factor affects the route.
My Role
This was my first time working with aviation weather and route planning, under a tight timeline. The project kicked off before product requirements were fully clear, and our PMs were still working with the airline team to align goals and expectations. To help move things forward, I proactively researched industry tools and AI weather models, and worked closely with stakeholders to define what the prototype needed to show.
I translated raw model outputs and high-level goals, such as multi-layer cloud forecasts and route planning scenarios into clear visual elements and interactive charts. These design decisions helped ground the team’s technical narrative, making it easier to communicate the model’s value to both internal teams and airline partners.