NASA Space Apps 2025 · Global Nominee

JWST Space Explorer

A full-stack explorer that retrieves public JWST observations from NASA MAST, processes FITS data into Deep Zoom imagery, and lets users inspect the results with an AI assistant.

Role
Team Lead · Developer
When
October 2025
Mode
Digital product
JWST Space Explorer landing page with a deep-space background and astronomical target search.
Search a target before retrieving and processing public JWST observations.

01 / Context

The problem

NASA’s “Embiggen Your Eyes!” challenge asked teams to make extremely large astronomical images accessible to the public. Raw FITS observations are difficult to retrieve, process, and inspect in an ordinary browser without specialist software.

Approach

The application separates a React exploration interface from a FastAPI processing service. A user searches for a target, the backend retrieves public observations from NASA MAST, converts FITS data into Deep Zoom tiles, and the browser presents the result through OpenSeadragon with an AI assistant for guided inspection.

02 / Contribution

What I contributed

  1. 01

    Led the team’s scope, priorities, and delivery through a 48-hour hackathon.

  2. 02

    Developed a Python and FastAPI workflow for querying NASA MAST and processing FITS observations.

  3. 03

    Prepared browser-ready Deep Zoom tiles with Astroquery, Astropy, and pyvips.

  4. 04

    Integrated OpenSeadragon for smooth inspection of large images and Gemini for guided analysis.

03 / Decisions

Technical decisions

01

Tile, don’t download

Convert large FITS observations into Deep Zoom tiles so the browser loads only the detail required at the current zoom level.

02

Separate processing from viewing

Keep NASA retrieval and scientific image processing in FastAPI while React and OpenSeadragon remain focused on exploration.

03

Assist the investigation

Use Gemini as a contextual assistant beside the scientific viewer instead of replacing the underlying observation data.

05 / Evidence

Outcome

  • Selected among 1,290+ Global Nominees from 11,500+ worldwide submissions.
  • Completed a working scientific-data experience inside the 48-hour event window.
  • Recognised by Le Quy Don High School for the Gifted as a notable student achievement.
Next system / 02AstroVerse↗