Staff Systems Architect, AI Infrastructure at Scale

Fireworks; San Mateo (CA); On-site; USD 180,000 - 240,000; Full time

Job summary

Fireworks is seeking a Member of Technical Staff for Systems Infrastructure to design and operate scalable infrastructure for AI model training and inference. You will work on schedulers, storage, and networks across thousands of accelerators in a production fleet.

The role emphasizes turning research ideas into production systems, with mentorship from senior engineers and flexible start dates around thesis defenses. Deep expertise in one focus area and strong systems programming are essential.

Qualifications

• PhD completed or expected by Dec 2026 in CS/CE/EE or related field.
• Research background in distributed systems, OS, scheduling, storage, networks, architecture, or HPC.
• Deep expertise in at least one focus area evidenced by dissertation/publications.
• Strong systems programming skills in C/C++, Rust, Go; proficient in Python.
• Experience building and evaluating real systems with measurable performance results.
• Ability to communicate complex systems design clearly.

Responsibilities

• Design, build, and operate core infrastructure for large-scale training and inference.
• Define benchmarks, traces, and simulators to reason about scheduling, caching, and network performance.
• Identify bottlenecks across kernel to scheduler, and drive improvements with data.
• Turn research ideas into production systems that withstand real workloads and failures.
• Collaborate with research and inference teams so infrastructure and model design align.
• Track emerging hardware, interconnects, and systems research to guide direction.

Skills

PhD

Distributed systems

Operating systems

Scheduling

Storage systems

Computer networks

Computer architecture

HPC

C/C++

Rust

Go

Python

Kubernetes

Cloud infra

Multi-tenant

Education

PhD in related field

Tools

Kubernetes

CUDA

ROCm

Triton

NCCL

Docker

Job description

Fireworks is seeking a Member of Technical Staff for Systems Infrastructure to design and operate scalable infrastructure for AI model training and inference. You will work on schedulers, storage, and networks across thousands of accelerators in a production fleet.

The role emphasizes turning research ideas into production systems, with mentorship from senior engineers and flexible start dates around thesis defenses. Deep expertise in one focus area and strong systems programming are essential.