Architected and developed a multi-region AWS Spot GPU finder/runner, using State
Machine, Batch, and Lambda
Implemented IaC methodology across the organization using Pulumi via 90% Python and 10%
TypeScript
Reduced overall AWS spend in EC2, S3, ECS, and RDS by ~20% with no impact on scientific
workloads
Designed AWS networking to support our on-prem GPU cluster, allowing constant processing
24/7 with reduced cost
Architected CryoSPARC using FSx for Lustre for high-throughput and autoscaling GPU nodes
via AWS ParallelCluster
Managed IT Ops and Security 4-person team responsible for on-prem and AWS
infrastructure, endpoint management, and cloud access controls
Architected Relay’s first Karmada-based Kubernetes platform to run GPU workloads across
EKS and other multi-cloud clusters via a unified control plane, simplifying job
submission for scientific teams
Optimized 50+ Docker containers, reducing image size, resulting in faster deployments
and lower resource usage
Managed IIS, Apache, and NGINX web servers for our internal and external applications
Managed Active Directory, Group Policy, and DNS for our AD domain
Managed Debian & Fedora Linux systems for our core web portal consisting of
WEB/APP/DB/NFS servers
Managed multiple VMware ESX 5.5 clusters with 300+ VMs
Maintained AWS cloud environment using EC2, VPC, S3, SES, and IAM features
Designed, installed, and maintained SharePoint 2013 with Enterprise features configured
and enabled
Created a web application for HR to update certain AD fields using ASP.NET and AngularJS
Created PowerShell scripts for automation of tasks
Created documentation and diagrams on new and existing applications & held meetings to
discuss
Managed Google Postini and assisted the migration to ProofPoint SPAM filters
Published applications and managed Citrix XenApp farm
Designed and managed connected SSO solutions using ADFS 3.0
Managed MobileIron MDM and responsible for compiling & deploying Apple iOS/Android apps
to public app stores to allow consumers access to latest app versions
Managed Cisco ACE load balancers for internal and external web applications
Managed storage on our EMC VNX 5500 SAN
Managed backups and restores using Simpana CommVault
Provided last level support for Developers, Helpdesk, QA & end users, resolving complex
problems
