Skip to content

NASA · 2017–2025

LAADS DAAC

Find Data search, ordering, and near-real-time access for NASA LAADS DAAC. The wizard is public. Web-system delivery runs through GitLab CI/CD and Kubernetes alongside existing Perl services.

Lead Software Engineer — web applications and delivery workflows.

  • Perl
  • Kubernetes
  • GitLab CI
On this page

Problem

  • Researchers need a clear path from choosing a product to finding and ordering files.
  • Web work has to keep working with the catalog, archive, and access services already in use.

Decisions

  • Build a guided search-and-order workflow organized around product, time, location, files, and review.
  • Treat the LAADS portal and near-real-time access tools as one data-access experience, not a single page.
  • Move builds and deployments to GitLab CI/CD, Docker, Helm, and Kubernetes without replacing the archive interfaces.

Outcome

  • Find Data is live — product, time, location, files, review and order. That is the public artifact.
  • Web-system delivery is more repeatable through GitLab CI/CD, Docker, Helm, and Kubernetes. Existing Perl services remained in place.
  • Portal and near-real-time access work sat in the same delivery path as Find Data. This page does not publish query-time or user-count metrics.

Team & contribution

Contribution
Technical leadership and web-system development
Team & partners
LAADS engineering, science operations, and user-support colleagues
What I improved
Search and ordering workflows, supported by a more repeatable release process.
Key decision
Modernize the user experience and delivery workflow while retaining the archive interfaces.

LAADS DAAC work covered the public portal, Find Data search and ordering, near-real-time access, and the delivery process around those web systems. The LAADS site is the broader public portal. Find Data remains the walkable artifact.

The underlying archive interfaces stayed in place. Modernization was the applications and how they shipped, not a rewrite of the archive.

Find Data

Find Data organizes a search into five steps: products, time, location, files, and review and order. That experience connects the user interface to the catalog, archive, and access workflows people already depend on.

A clearer interface still has to complete an order against those services. The work could not stop at the page design or ignore existing data-access paths.

The same delivery path included the LAADS portal and its near-real-time tools. They were connected parts of one access experience, not a standalone microsite.

Delivery around existing services

The environment included established Perl services. Alongside application development, builds and deployments moved into GitLab CI/CD, Docker, Helm, and Kubernetes.

The emphasis was repeatable delivery while retaining the interfaces to the archive. Incremental modernization of the web systems — not a claim that every service behind LAADS was replaced.