I led multiple components of a larger project within Palantir, including deployment of Palantir's Gotham platform on OpenShift on-premises, and the creation and documentation of an SDK for interacting with Palantir's TITAN software.

I led a three person team responsible for maintaining the performance, stability, and security of multiple instances of Palantir's Foundry and Gotham platforms at a Swiss client site.

My team managed the largest on-premises Foundry instances in relation to user scale and platform activity, and as such we navigated performance and stability challenges not experienced by smaller on-premises deployments or by cloud-hosted instances. This required collaboration with client IT staff and Palantir development teams, a deep understanding of Palantir software components and adjacent open-source technology, and precise execution of complex operational tasks on production systems, i.e., database migrations and hardware migrations.

I was the subject matter expert for Apache Hadoop (primarily HDFS and YARN) within Palantir. This entailed performance and stability debugging of interactions between Hadoop components and Palantir libraries and services, maintaining an internal fork of Hadoop, and engaging with the open source community by submitting patches and commenting on proposed changes to YARN scheduling.

I would also travel to customer sites with particularly complex or challenging interactions between Palantir software and Hadoop components, including performance issues, hardware migrations, and compatibility issues with externally managed Hadoop clusters.