Director, Generative AI Platform & Systems

Capital One; McLean (VA); On-site; USD 150,000 - 190,000; Full time

Job summary

Capital One is seeking a leader to scale and govern our AI training platform, enabling foundation models at enterprise scale. You will oversee distributed training, evaluation, and deployment workflows, with a focus on cost efficiency and reliability.

The role works cross-functionally with researchers, engineers, and product teams to deliver AI-powered capabilities that improve customer experiences and banking processes.

Qualifications

• Experience building large-scale distributed AI training systems.
• Experience with GPU scheduling and fault tolerance.
• Familiarity with Open Source AI tools and workflows.
• Ability to design scalable, cost-aware architectures.

Responsibilities

• Partner with engineers, scientists, PMs to deliver AI-powered products.
• Oversee design, development, testing, deployment of core AI platform systems.
• Make build-vs-buy decisions across OSS and SaaS AI tech like PyTorch/HuggingFace.
• Invent state-of-the-art techniques to improve scalability and reliability.
• Own GPU capacity planning and cost governance for training workloads.
• Contribute to long-term AI systems Roadmap and platform vision.
• Attract and grow AI talent; foster a culture of learning and governance.

Skills

Distributed systems

GPU scheduling

Open Source AI

Cost governance

Tools

AWS Ultraclusters

HuggingFace

PyTorch

VectorDBs

Job description

Capital One is seeking a leader to scale and govern our AI training platform, enabling foundation models at enterprise scale. You will oversee distributed training, evaluation, and deployment workflows, with a focus on cost efficiency and reliability.

The role works cross-functionally with researchers, engineers, and product teams to deliver AI-powered capabilities that improve customer experiences and banking processes.