Senior Staff AI/ML Infrastructure Engineer building automation, validation, and agentic engineering systems across GPU platforms, HPC, and cloud infrastructure.
Own the architecture and development of automated validation capabilities for AI infrastructure deployments and post-repair verification, shifting engineering effort toward exception handling.
Designed and built an internal, multi-step agentic coding workflow with QA stages, creating a governed engineering workflow with less routine human intervention.
Led AI infrastructure validation decisions for several data-center builds of 100 MW or greater, aligning technical execution across delivery, operations, and engineering stakeholders.
Began development of automated validation orchestration to improve operational efficiency, reduce validation time, and reduce manual effort.
Continued the GPU VM-host modernization program, mentored new engineers, and developed an Ansible-based GPU validation framework.
Stabilized Cloud GPU outages within 60 days of joining and rebuilt the Cloud GPU/vGPU orchestration product to production quality using NVML bindings.
Improved vGPU-license efficiency and led a cross-team migration of GPU VM hosts from Debian 10 to NVIDIA-supported Ubuntu 24.04 hypervisors.
Led up to 13 developers, hiring and mentoring engineers while establishing development processes and Agile practices; reported directly to the CEO.
Joined customer POC and go-to-market conversations, and led adoption of GPU and AI-infrastructure cluster capabilities, including E4S and other AI/ML tool stacks, into ODDC.
Maintained cloud-hosted HPC images and built a CI/CD validation pipeline; mentored developers, performed code reviews, and began release-engineering responsibilities.
Served as a developer point of contact in customer conversations, connecting product feedback with engineering execution.
Added BeeGFS and NVIDIA Multi-Instance GPU support to HPC infrastructure capabilities.
Achieved 99.999% uptime and increased peak traffic capacity by 40% by engineering a highly available WordPress cluster using Terraform, Ansible, and AWS.
Reduced page load times by 50% and managed a 100% increase in high-traffic demands by designing a high-performance WordPress tech stack with OpenLiteSpeed, Redis, and FreeBSD.
Enhanced application functionality by developing and maintaining a Node.js application, ensuring seamless integration with external vendor APIs.
Streamlined VOIP services across 200+ US hotels by developing a Debian Linux-based image and standardizing configuration with Ansible, significantly enhancing system reliability and efficiency.
Drastically reduced installation times and labor costs by automating equipment installation configurations, cutting onsite install time by 67% and pre-installation man hours by over 80%.
Mitigated service disruptions by serving as Linux/Network SME and implementing root cause analyses and post-mortem meetings following service outages, effectively enhancing system resilience.
Enhanced WordPress site performance, achieving a 50% improvement in page load times and better SEO rankings through custom theme and plugin modifications.
Streamlined content updates for non-technical team members by integrating advanced custom fields in WordPress, reducing developer dependency by 70%.
Boosted user engagement by 40% and increased page views per visit by 30% with interactive WordPress features like custom post types and user forums.
Optimized delivery workflows by orchestrating build and release processes in Java environments using Jenkins/Hudson, ensuring timely and error-free deployments.
Enhanced development efficiency by streamlining CI/CD pipelines in collaboration with developers, accelerating build times by 25%.
Boosted customer satisfaction by enhancing software release quality in partnership with QA teams, implementing rigorous quality assurance practices.
people have visited this page.