Relevant Experience & Education Highlights

Demonstrates deep expertise in high-performance ML infrastructures, GPU optimization, and generative AI capacity planning, making a strong fit for building scalable foundation model systems at Adobe.

1. Build and optimize infrastructures that power large foundation model training on thousands of GPUs

Led development of hyperscale ML systems integrating hardware and software for generative AI workloads.

• Led technical initiatives in ML systems and cloud organization at a major search and cloud company, designing low-level computer systems interacting with kernel and hardware for ML use cases.
• Oversaw end-to-end hardware infrastructure for generative AI at a leading social media company, managing ordering, new product introduction, and fulfillment processes.
• Designed and optimized HPC and generative AI infrastructure at a semiconductor industry leader, leading team to ensure scalable and cost-efficient resource utilization.
• Built GPU infrastructure and accelerated computing pipelines at a major telecommunications company, automating GPU allocation, testing, and certification processes.

2. Profile GPU utilization, trace inference and training runs and help craft strategies for optimizing our ML model latency

Specialized in performance engineering and GPU characterization to resolve bottlenecks and enhance efficiency.

• Conducted performance analysis and optimization for AI workloads on hardware at a semiconductor industry leader, focusing on benchmarking, code optimization, and power engineering.
• Led identification and resolution of performance bottlenecks across platforms at a major telecommunications company, implementing large-scale testing and profiling strategies.
• Orchestrated performance improvements such as TLB shootdowns and MobileMark workload studies in system power and performance roles.

3. Architect and optimize end-to-end ML pipelines, ensuring they're scalable, efficient, and robust

Drove capacity planning, systems integration, and cross-functional collaboration for AI-driven solutions.

• Led capacity planning for generative AI models at a leading social media company, developing, optimizing, and scaling systems powering diverse products.
• Partnered across ML, hardware, SRE, and cloud teams at a major search and cloud company using C++ and Go for full-stack systems including cluster management and Linux kernel.
• Architected AI-driven solutions for performance analysis optimizing hardware handling of AI models, algorithms, and workloads.

Requirements & Candidate Alignment