Apply on Employer Site

Advanced Microdevices Pvt. Ltd. (India) · 7 hours ago

CVP FDE – AI Software Development

San Jose, CA

Full-time

Onsite

Director/Executive

Advanced Micro Devices, Inc is dedicated to building great products that accelerate next-generation computing experiences. The role focuses on leading a world-class FDE organization, ensuring high-velocity customer outcomes, and optimizing GPU cluster performance for customer success.

BiopharmaBiotechnologyIndustrialManufacturing

No H1B

Responsibilities

Build and scale a world-class FDE organization, strategically combining ML Generalists, Low-Level Kernel Optimizers, and Solutions Architects to cover the full customer deployment lifecycle

Define and institutionalize the FDE Engagement Model to maximize resource leverage and ensure consistent, high-velocity customer outcomes

Serve as the Voice of the Customer internally: Translate field intelligence and customer challenges into concrete, prioritized engineering roadmaps, and ensure execution

Cluster Bring-up & Optimization: Oversee the technical onboarding of massive GPU clusters. Ensure your team can troubleshoot collective communication errors, debug framework issues, and optimize training/inference strategies

Utilization Engineering (The North Star Metric): Drive and maintain industry-leading Customer GPU Utilization across clusters of thousands of GPUs, making cluster satisfaction the key measure of success

High-Performance Model Deployment: Enable customer success by deeply optimizing open-sourced models (Llama 3, DeepSeek, Mixtral) and proprietary models for our specific hardware topology, utilizing tools like vLLM and TensorRT-LLM

Executive Technical Sponsorship: Act as the technical authority and executive sponsor on large deals, possessing the credibility to validate architecture with CTOs and VPs of AI

Feedback Loop: Aggressively channel field intelligence back to Product Engineering. If customers are struggling with a specific use case, then become the loudest voice in the room demanding a fix

Qualification

AI FrameworksDistributed ComputingGPU EcosystemLLM OperationsTechnical LeadershipCommercial AcumenCrisis ManagementHybrid Hardware/Software

Required

BS, MS or equivalent with direct experience

Preferred

Years of Technical Leadership: Demonstrated track record leading high-impact technical teams within high-stakes environments (e.g., Cloud Infrastructure, AI Platform, or HPC)

The 'Hardware/Software' Hybrid: You understand the stack from the metal up

Commercial Acumen & Fluency: Deep understanding of commercial drivers (ARR, Churn, Margin) and the ability to articulate how technical solutions impact deal velocity and business outcomes

High-Stakes Crisis Management: Experience leading through 'Sev0' customer incidents (e.g., massive training run failures), demonstrating the poise and clarity required to manage executive communication while guiding rapid root cause resolution

AI Frameworks: PyTorch, JAX, TensorFlow

Distributed Computing: Slurm, Ray, Kubernetes (K8s), Docker

GPU Ecosystem: NVIDIA drivers, CUDA profiling (Nsight Systems), Triton Inference Server

LLM Operations (Differentiator): Significant experience with advanced LLM deployment and customization techniques, including fine-tuning (e.g., LoRA/QLoRA) and building RAG pipelines