Product / Palantir.com

Palantir Analytics

Led a major systems-design project redesigning Quiver, turning a niche time series analysis tool in Palantir's Foundry platform into a widely used low-code visual programming tool and empowering future LLM-based analytical workflows

Context

The Analytics suite is a product organization within Palantir that focuses on leveraging an organization's "Ontology" (an organization's digital data win, modeled as data objects with semantic relationships). It's used to build scalable, sophisticated workflows with ease and stability.

Quiver originally developed as two separate apps that evolved from different use cases and technical histories: one custom-developed by end-users to analyze and monitor time series data from high-scale sensors, and one dashboard creation tool that allowed users to leverage high-quality, organization-approved datasets (defined as “objects” in a client’s data ontology) to perform calculations and create data visualizations. In contrast to Palantir's other coding-based analytics tools, Quiver offered a point-and-click interface that required minimal to no coding.

Impact

Over 2 years, I redesigned Quiver to convert it from a niche time series tool to a more generalized low-code analysis tool. This allowed less technical users more easily create dashboards and analyses with various input data formats, especially time series.

This was a systems design-focused project that required going back to first principles: we conducted extensive user research on a high variety of use cases, decomposed and synthesized these disparate use cases into a singular vision, and proposed a series of high-lift product changes that completely shifted the position of the app in Foundry’s ecosystem to be generalized to cover a wider variety of use cases than its more specific origins.

This vital repositioning of the product generalized its usefulness, and my design system created a foundation of primitives that enabled the app to be transformed with LLM features in the tail end of the project. It has been used by business analysts and stakeholders for historical analyses, live monitoring dashboards, and future scenario planning. It's made headlines for aiding in anti-money laundering efforts at major world banks, streamlining workflows at UK's National Health System, facilitating the conversion to sustainable energy sources at a supermajor, and more.


Please contact me with any inquiries about this project. All data is notional and mocks shown here are representational.

Analytical nodes are laid out in a graph that displays data lineage.