AI Data Infrastructure Engineer — Storage & Caching

Fireworks; San Mateo (CA); On-site; USD 180,000 - 280,000; Full time

Job summary

Fireworks seeks a senior storage/data-infra engineer to own the data layer across multi-cloud environments for AI workloads. You will design distributed storage systems, caching layers, and high-performance data paths that keep datasets and model weights in the right place at the right time.

Responsibilities include optimizing replication, data movement, and coherence; deploying clusters with Terraform or Pulumi; and profiling bottlenecks across kernel, network, and storage backends.

Qualifications

• Bachelor’s degree in Computer Science, Engineering, or related field or equivalent practical experience.
• 5+ years designing, building, and operating storage, caching, or data infrastructure at scale.
• Strong software development in C/C++, Rust or Go; Python/Go for automation.
• Deep understanding of distributed storage: consistency, replication, and metadata scaling.
• Hands-on Kubernetes (CSI) to mount high-performance volumes into ephemer-GPU pods.
• Experience deploying storage clusters across clouds with Terraform or Pulumi.
• Experience with benchmarking and profiling tools (fio, iperf3, NCCL, perf, eBPF).

Responsibilities

• Design and build a unified data layer across clouds for AI workloads.
• Own storage, caching, and data movement for training and inference.
• Optimize placement, replication, and movement of datasets and weights.
• Collaborate with GPU teams to ensure low-latency I/O and scalable KV storage.

Skills

Storage design

Distributed caching

C/C++

Rust

Go

Python scripting

Kubernetes CSI

Terraform

Pulumi

Performance profiling

Raft consensus

Education

Bachelor's degree in CS/Engineering

Tools

Kubernetes

Ceph

Lustre

MinIO

WEKA

VAST

fio

iperf3

NCCL

eBPF

io_uring

Job description

Fireworks seeks a senior storage/data-infra engineer to own the data layer across multi-cloud environments for AI workloads. You will design distributed storage systems, caching layers, and high-performance data paths that keep datasets and model weights in the right place at the right time.

Responsibilities include optimizing replication, data movement, and coherence; deploying clusters with Terraform or Pulumi; and profiling bottlenecks across kernel, network, and storage backends.