AI Datacenter Systems
NVIDIA AI infrastructure, AI factory concepts, GPU platforms, rack-scale design, workload readiness, cluster operations, and TCO optimization.
Datacenter AI Infrastructure | Storage | Linux Systems
I help customers and partners design, validate, and launch datacenter platforms across AI/ML, storage, networking, Linux system software, and rack-scale infrastructure.
Experienced cloud hardware and software solution architecture professional with 20+ years across datacenter, storage, networking, and AI/ML. Proven track record designing and implementing scalable cloud solutions that improve performance, reduce TCO, and support business growth. Strong partner-facing leadership across engineering, architecture, program, and customer teams, with hands-on Linux debug experience and a record of moving complex projects from concept through validation and launch.
NVIDIA AI infrastructure, AI factory concepts, GPU platforms, rack-scale design, workload readiness, cluster operations, and TCO optimization.
NVMe, NVMe-oF, SPDK, RAID, OpenZFS, RDMA, TCP/IP, VLAN/QoS, persistent memory, and high-performance cloud storage architecture.
Linux system software, drivers, firmware/BMC management concepts, Redfish, PLDM/RDE, telemetry, CI/CD, Jenkins, Ansible, and Kubernetes.
Lead Enterprise and AI datacenter solution enablement, including NVIDIA AI factory and agentic AI infrastructure proposals, qualification, stakeholder alignment, launch readiness, and customer adoption.
Led CSP technical engagements across data, storage, and AI/ML solutions on Xeon, NVMe SSD, and Persistent Memory platforms, including petabyte-scale cloud block storage design wins.
Managed NVMe SSD system software solutions for 15+ OEM and CSP server platforms, enabling design-in, integration, validation, release readiness, and production adoption.
Led Linux open-source stack enablement with OEM partners across Android tablets, netbooks, and IVI platforms, improving release cadence and supporting millions of shipped devices.
Built embedded, Linux, RTOS, networking, security, image processing, and SoC platform software while leading customer-facing development and debug engagements.
Optimized approximate nearest neighbor workloads for terabyte-scale datasets, fitting larger working sets into a single server while reducing TCO by 30% versus DRAM-based approaches.
Designed CSP block storage using petabyte-scale QLC drives, SPDK FTL, and Persistent Memory buffering to reduce OPEX by 40% while addressing endurance and performance bottlenecks.
Improved ZFS synchronous write and read cache performance using Persistent Memory as ZIL/SLOG and L2ARC, improving random write performance and reducing read latency by 98%.
Enabled Intel VROC and VMD solutions for scalable NVMe RAID, error isolation, LED management, bootable RAID, and surprise hot-plug stability across Linux, Windows, and HCI environments.
Led complex cloud and datacenter solution engagements across multi-geo teams, partners, and customers, delivering multimillion-dollar business value. Known for turning ambiguous technical requirements into practical product plans, issue triage rhythms, customer-ready architecture decisions, and measurable launch outcomes.