Senior AI Compute & Data Infrastructure Architect

Vinci; Palo Alto (CA); On-site; USD 180,000 - 240,000; Full time

Job summary

Vinci is building a world-class foundation for physics AI training in the Palo Alto area. The engineer will own data and compute infrastructure, enabling rapid experimentation and reliable model training at scale. You will lead design choices, write production code, and mentor a growing team.

The role focuses on architecture, cost-aware optimizations, and ensuring data availability for the AI team’s ambitious research agenda. Join a small, technically focused group driving scalable performance.

Qualifications

• 10+ years building large-scale distributed systems.
• Direct experience operating multi-node distributed training on GPUs.
• Experience serving training data at petabyte scale.
• Experience setting scheduling and quota policy on shared GPU capacity.

Responsibilities

• Define how GPU capacity is organized and grown as the fleet expands.
• Own workload scheduling to manage training, validation, and data prep tasks.
• Design storage, ingestion, and access paths for scalable training data.
• Own the systems to launch, checkpoint, resume, and monitor runs.

Skills

Distributed systems

GPU infrastructure

Multi-node training

Scheduling policies

Data engineering

Mentoring

Technical leadership

Problem solving

Education

Advanced degree in CS or related field

Tools

Kubernetes

Slurm

Ray

Airflow

DeepSpeed

PyTorch

Job description

Vinci is building a world-class foundation for physics AI training in the Palo Alto area. The engineer will own data and compute infrastructure, enabling rapid experimentation and reliable model training at scale. You will lead design choices, write production code, and mentor a growing team.

The role focuses on architecture, cost-aware optimizations, and ensuring data availability for the AI team’s ambitious research agenda. Join a small, technically focused group driving scalable performance.