SimSpace Weaver is a managed service for scaling spatial simulations/games. At launch, Weaver was capable of running running 2M+ dynamic entities at 30hz. I led the design and technical efforts of a team of 4-6 engineers. The core data plane technology is an in-house spatial database built on a log structured, partitioned, real-time, shared-memory storage engine with tunable consistency (all written in Rust!). Customers use Weaver by integrating directly with Weaver App SDK or using the Unreal or Unity game engine plugins.

Short for "doppelganger," dopl aims to provide concise and pertinent recommendations to users by correlating their individual desires and providing uniqueexperiences by disregarding the ineffective loud-sourcing. Instead, dopl uses machine learning and behavioral analytics to provide consumers with pertinent recommendations that are based on their preferences, and most importantly, their behavior.

● Collaborator in the design and implementation of the cluster infrastructure software for the distributed log based data store for Amazon Aurora (Java)● Designed the health and metrics monitoring support for data storage nodes● Participated in on ­call rotations for operational support of customer facing issues● Developed and maintained of the team wide operations framework/toolkit (Python)● Developed a pluggable service discovery library as a replacement for DNS● Developed the first generation of software deployment infrastructure● Developed a holistic integration test and control framework for testing a distributed service oriented infrastructure containing multiple heterogeneous web services across multiple data centers and clusters● Implemented the authentication, authorization, and encryption strategies for external facing API surfaces● Spearheaded the redevelopment, and took over maintenance of, the cluster building and repair framework● Designed and developed the infrastructure for managing database instance groups (one writer, multiple readers) life­ cycles

● Investigated the practicality and implementation of a modular, hierarchical, group scheduling framework as a viable replacement for Priority Inheritance Scheduler in the Linux RT Kernel.● Created new scheduling kernel modules for real time and embedded applications.● Developed Python and C libraries for user level control kernel control.● Designed user level tests for module testing,● Utilized kernel and user application instrumentation for debugging.● Sampled, analyzed, and visualized test data from various test execution scenarios.

● Led in the development of a Django (Python) based framework for compiling comparative analyses of legal documents.● Investigated different metrics for graphical representations for legal documents.● Developed a meta search engine for legal research that aggregated automated search results from free and subscription services (LexisNexis, Westlaw, FindLaw, etc.).● Researched the use of a distributed RPC processing pipeline for the agglomerated legal documents.