Supported multiple teams with day-to-day execution of tasks, mentoring/onboarding, technical direction, and strategic vision to verify the AI driving system's safety metrics.Led quarterly, milestone, and roadmapping planning for teams and cross-functionally. Developed technical strategies and assembled working groups to meet milestone deliverables and other priority needs.Zoox Safety Analysis Framework: Planned and executed team's contributions to two milestones using systematic simulation to assess AI driving system safety, including helping to develop the overall Zoox safety strategy. Restructured team to better align with milestone deliverables, and balanced work between near-term deliverables, developing infrastructure, and long-term R&D. Stood up cross-functional working groups to apply data science to driving data to qualify safety requirements and integrate the Zoox-wide safety argument.Zoox Collision Avoidance Test Suite: Developed quarterly and milestone planning to get the project started from scratch and supporting driving safety analysis in less than 3 months. Tracked progress, identifying and resolving potential blockers prior to them becoming issues. Provided technical guidance and mentorship to team members, enabling them to meet milestone deliverables.Vehicle Verification & Validation Testing: Developed frameworks to validate human operator performance and simulation validation to demonstrate safety requirements were met. Pulled together a cross-functional team to develop operator training and supported the team in successfully executing the work to meet milestones. Creating an integrated plan to efficiently validate simulation tooling used to demonstrate AI driving performance.Cross-Functional Working Groups: Initiated and stood up cross-functional working groups to close critical capability gaps:* Comparative driving analytics* Validation of predictive tools* Human behavioral modeling process* Safety metrics infrastructure

Responsibilities in the areas of strategy, operations, team leadership, project development, budgeting, com- munications, and stakeholder engagement for a $200M/yr Department of Energy (DOE) high-performance computing program. Engaged partners across Sandia and DOE to deliver computing platforms and simulation/data analytics software in support of DOE’s mission.• Strategy:Led development of machine learning strategy for the Sandia ASC program. Strategy was presented to the DOE, leading to follow-on federal-level strategy development and preliminary funding. Connected ML strategy to Sandia’s high-performance computer acquisition strategy, being used at the DOE to drive future platform procurements. Spearheaded a high-performance computing strategic partnership with a DOE manufacturing agency, identifying and roadmapping areas of mutual collaboration. New funding obtained for two consecutive years to support maturation of the collaboration. Currently leading an AI-enabled model repository development effort. Supported program-level strategic planning, leading to definition of 5 year strategic program targets.• Stakeholder Engagement:Initiated and matured connections between the ASC program and stakeholders across Sandia, including representing the program in cross-cutting groups to improve partnering and coordination between different programs. Used elicitation strategies to understand and incorporate stakeholder needs into ASC plans. Lead program technical communication to DOE stakeholders through written summaries, planning documents, and organizing technical seminars.• Operations:Lead the program’s operations team to determine high-performance computing policies, resource allocation, and priority access. Liaised with partner laboratories to ensure DOE-level resources will productively support Sandia user’s needs. Additional responsibilities include budgeting and reporting.- Served as Sandia Verification & Validation Program Element lead (acting) and deputy.

In this position, I have been the PI for the following R&D efforts:1) Atomistic-to-continuum (AtC) modeling for solid/fluid interfaces at the nanoscale,2) Creation and use of data analytics, uncertainty quantification, and verification & validation for heat transfer models.3) Atomistic modeling of electric double layers.4) Calibration and uncertainty quantification of large-eddy simulations for turbulent combustion.5) Developing and implementing a mesh free modeling capability.6) Creating a Bayesian network model to assess large engineering projects.In addition, I was a team member on the following projects:1) Creating Machine Learning strategies to identify and correct errors in turbulence models.2) Developing crystal plasticity variability models using Bayesian estimation and Machine Learning.3) Developing a computational capability to model additive manufacturing processes.

I was the PI of two 6-month research projects:1) Development of a finite element multiscale electric field solver for molecular dynamics,2) Development of coupling techniques for in situ wind farm modelingand led the model development and quantification of margins and uncertainties effort for thermal loading of a large scale engineered system.I also was a team member on the following projects1) Development and implementation of the theory and algorithms for the atomistic-to-continuum (AtC) package to couple thermal, mechanical, and electron transport physics,2) Implementation of the SST turbulence model in Sandia's in-house CFD software.

I contributed to two projects during this time:1) Modeling, simulation, and design of a micro-fluidic mixer using electro-kinetic flows,2) Developing atomistic-to-continuum modeling theory and software for heat transfer in solids.

I developed the theory and software for large-eddy simulation (LES) wall models based on explicit filtering. This built on my thesis work in which I developed and implemented two different LES wall models, one based on optimal control theory and the other on viscosity stress matching, in a code I wrote from scratch.