As a Principal Solutions Architect supporting Foundation Model Providers (FMP) on AWS, you will tackle some of the most challenging and exciting problems in cloud computing today. Model provider's compute footprint represents one of the most demanding workloads on the AWS platform, pushing the boundaries of what's possible with networking, GPU infrastructure, storage, container orchestration, and distributed computing at extraordinary scale.
In this role, you'll design cloud architectures that enable model providers to train, fine-tune, and serve state-of-the-art generative AI models. You'll help solve technical challenges that few organizations in the world face - exabyte-scale data, millions of interconnected GPUs, complex networking topologies, and custom hardware acceleration requirements that define the leading edge of AI infrastructure. You will help them solve business challenges such as rapidly releasing products/services to the market and building elastic, scalable, cost optimized applications. You will engage with product owners and service teams to set the strategy for AWS services. For this role, we are looking for folks who have technical breadth complimented by technical depth in one or two areas, business aptitude, and the ability to lead in-depth technology discussions, articulating the business value of the AWS platform and services.
Key job responsibilities
• Maintain and foster relationships with model providers, becoming their trusted technical advisor and strategic partner
• Develop deep knowledge of core foundational services (Compute, Network, Storage) along with ML expertise to build long-term relationships with customer engineering teams
• Dive deep to understand the details of model provider's environment, business goals, and technical requirements for building and deploying foundation models
• Design and implement advanced cloud architectures that enable model providers to scale their AI research and production workloads efficiently
• Partner closely with AWS service teams (EC2, Global Networking, EKS, Bedrock, S3) to influence roadmaps and develop custom solutions that meet model provider's unique requirements
• Identify patterns and technical solutions that can be broadly applied across the FMP segment to accelerate innovation
• Lead technical discussions that articulate the business value of AWS platform and services to both technical architects and executive stakeholders
• Drive technical and architectural best practices for GPU optimization, network throughput, distributed training, and cost efficiency at massive scale
BASIC QUALIFICATIONS
- 10+ years of specific technology domain areas (e.g. software development, cloud computing, systems engineering, infrastructure, security, networking, data & analytics) experience
- Bachelor's degree in computer science, engineering, mathematics or equivalent
- Experience developing technology solutions and evangelising end-to-end technology roadmaps that guide IT transformations toward cloud computing
- Experience communicating across technical and non-technical audiences and at C-level, including training, workshops, publications
PREFERRED QUALIFICATIONS
- Knowledge of large scale automation and workflow management or equivalent
- Knowledge of presentations and whiteboarding skills with a high degree of comfort speaking with internal and external executives, IT management, and developers
- Experience with training and deploying machine learning systems to solve large-scale optimizations, or experience operating highly available, distributed systems of data extraction, ingestion, and processing of large data sets
- Experience with CUDA kernels or ML/low-level kernels, or experience with Machine Learning and Large Language Model fundamentals, including architecture, training/inference lifecycles, and optimization of model execution
- Experience in Kubernetes, Docker or containers ecosystem
- Knowledge of foundation model architectures, training approaches, and serving infrastructure
