Member of Technical Staff, Data Infrastructure Storage
Fireworks AI2 days ago
Base Salary
$175k - $220k/yr
Responsibilities
- Design and build a unified, high-performance data layer for GPU training and inference workloads.
- Store, cache, replicate, and move datasets, checkpoints, model weights, and KV cache across clouds and storage tiers.
- Develop performance-critical data-path software in C/C++, Rust, or Go, with Python or Go automation where appropriate.
- Address distributed storage concerns including consistency models, replication, consensus, metadata scaling, caching, compaction, and failure recovery.
- Deploy and operate storage clusters and mount high-performance volumes into ephemeral GPU pods using Kubernetes and CSI.
- Benchmark and profile kernel, network, cache, and storage performance to identify and resolve bottlenecks.
- Support high-performance networking and data movement using RDMA, RoCE v2, InfiniBand, and GPUDirect Storage.
Requirements
- Bachelor’s degree in Computer Science, Engineering, or a related technical field, or equivalent practical experience.
- At least 5 years of experience designing, building, and operating production-scale storage, caching, or data infrastructure.
- Strong software development skills in C/C++, Rust, or Go, including performance-critical data-path code.
- Deep understanding of distributed storage fundamentals, including consistency models, replication, consensus, metadata scaling, caching, and failure recovery.
- Hands-on Kubernetes experience, including CSI, for high-performance volumes and ephemeral GPU pods.
- Experience deploying and operating storage clusters such as WEKA, VAST, Lustre, Ceph, or MinIO across clouds with Terraform or Pulumi.
- Proficiency with performance benchmarking and profiling tools such as fio, iperf3, NCCL tests, perf, and eBPF.
- Preferred experience contributing to or owning storage, caching, or key-value systems such as Ceph, Alluxio, JuiceFS, RocksDB, FoundationDB, TiKV, Redis, Dragonfly, SeaweedFS, or MinIO.
- Preferred experience with HPC or large-scale AI clusters, Slurm, Kubernetes GPU pools, AI I/O patterns, Linux internals, high-performance networking, GPUDirect Storage, and petabyte-scale cross-cloud data movement.
- Preference for experience with software-defined storage rather than proprietary hardware appliances.
Benefits
- Work on challenging AI infrastructure problems involving low-latency inference and scalable model serving.
- Collaborate with engineers and AI researchers in a fast-growing, inclusive environment.
- The posting describes ownership, direct impact, limited bureaucracy, and access to cutting-edge technology.
Tech Stack
Categories
About Fireworks AI
Fireworks AI builds a generative AI platform for developers and enterprises to train, fine-tune, and serve open models for production use across text, image, audio, embeddings, and multimodal workloads. It offers managed inference and tooling via APIs on globally distributed infrastructure, with a usage-based SaaS model. Founded in 2022 and headquartered in San Mateo, CA, Fireworks AI is a privately held, Series D company.