Member of Technical Staff, AI Cloud Storage
Location: Bay Area/ Seattle/ Remote
Reports to: CTO
About Neo Cloud
Neo Cloud is a fast growing next-generation AI neocloud, founded by ex NVIDIA,
CoreWeave and Intel engineering leaders. We offer you the opportunity to build
and operate next generation AI infrastructure that powers the world's most
demanding AI workloads for the largest AI labs. We are rethinking how to run
large scale AI infra for instance by leveraging agentic AI to build digital
twins to de-risk our physical deployments. Our agents are deploying and
monitoring our fleet to maximize the uptime of our infra.
About the Role
We're looking for a Principal Software Engineer to define and build Neo Cloud's
AI cloud storage platform. This is a senior individual-contributor role for an
experienced system architect who can operate at the intersection of customer
needs, system architecture, and production operations — translating what AI/ML
customers will need in the future, into a storage system that is performant,
resilient, and economical at massive scale.
Key Responsibilities
Identifying customer requirements:
Engage directly with customers, solutions architects, and product teams to
understand future storage requirements for AI/ML workloads.
Extrapolate from current usage patterns and industry trends to anticipate
future requirements.
Partner with product management to prioritize platform investments based on
near-term customer needs and longer-term strategic bets.
System design, implementation, and operations:
Design, implement, and operate AI cloud object and file storage systems,
including the data path, metadata path, and control plane.
Take a system-level approach that accounts for the full characteristics of AI
workloads, building end-to-end solutions — including local node caching
strategies, the network hardware and protocols that move data between storage
and compute (e.g., RDMA, high-throughput NICs, congestion control), and the
underlying storage hardware and software (media, erasure coding, metadata
services) — and understanding how decisions in one layer constrain or unlock
the others.
Architect for the specific demands of AI workloads: very high aggregate
throughput to keep GPU/accelerator clusters fed, support for massive numbers
of small and large objects and files, efficient checkpointing at scale, and
predictable tail latency under heavy concurrent load.
Drive core storage system design decisions, including durability and
consistency models, erasure coding and replication strategies, metadata
scalability, multi-tenancy and isolation, and S3-compatible and
POSIX/file-protocol API design.
Take end-to-end ownership of services in production: build for observability
and operability from day one, participate in on-call, lead incident response
and root-cause analysis for critical issues, and drive long-term reliability
and performance improvements.
Identify and eliminate performance bottlenecks and scalability limits before
they become customer-facing problems; lead capacity planning for rapid
growth.
Deep understanding of data privacy and security and its implication on
performance.
Ability to partner with network engineers to deliver complete AI storage
system.
Technical leadership:
Strong bias to action and resolution of technical decisions.
Set technical direction and best practices for the storage organization;
author and review design documents for significant architectural changes.
Provide deep technical mentorship to senior and staff engineers; raise the
engineering bar across the team through code review, design review, and
hands-on collaboration.
Influence technical strategy across adjacent teams.
Qualifications
10+ years of professional software engineering experience building and
operating cloud storage systems in production.
Direct experience with AI-focused storage platforms such as DDN, Weka, or
VAST, including their architectural approaches to throughput, caching, and
GPU-cluster integration.
Deep understanding of distributed systems fundamentals: consistency models,
replication, consensus, partitioning, failure detection, and recovery.
Proven experience operating high-scale distributed systems in production,
including on-call ownership, incident response, and driving systemic
reliability improvements.
Strong systems programming skills (e.g., Go, C++, Rust, or Java) and comfort
working across the stack from low-level I/O and networking to distributed
control planes.
Excellent written and verbal communication skills.
Nice to have
Experience designing or tuning local node caching layers to accelerate AI
training and inference data access.
Experience with high-performance networking.
Contributions to open-source storage projects, relevant patents, or published
technical papers/talks.