Member of Technical Staff - Cloud / HPC / Systems

Exclusive - Client is a startup AI Systematic trading firm (Founder is ex top HFT business leader)

Location: Singapore

Nationality: Singaporean is a must

Only open for Tech firm talents

Duties:

• Own and build core infrastructure: Slurm cluster on Spot instances, OpenZFS filesystem, S3 for market data, node-local NVMe scratch, and researcher workstations
• Write and deploy the Python/shell that runs the research data pipeline — ingestion, normalisation, partitioning, job submission, checkpointing, reconciliation
• Build agentic workflows using coding agents (Claude Code) to automate real steps in the platform
• Support researchers directly — debug failed jobs, diagnose root causes, and change the platform so failures don't recur
• Evaluate technical options (build vs. buy) and write comparisons the founder can act on
• Over time, take ownership of firm-wide data strategy and compute cost management

Requirements:

• Real production experience running workloads on Spot/preemptible compute — not just familiarity
• Deep Linux systems knowledge: filesystems, processes, memory, scheduling
• Practical experience with ZFS/OpenZFS, NFS, local vs. network storage
• HPC/cluster background a strong plus: Slurm, PBS, LSF, MPI, or on-prem scientific computing
• Strong Python; comfort with shell/systems glue; enough fluency in a compiled language (Rust, C++, Go, Java) to read and profile code
• Experience with cost optimization, networking (VPC, DNS, cross-AZ/region transfer), and security fundamentals (IAM, least privilege, secrets management)
• Track record of owning production incidents end-to-end
• Terraform/CDK and CI/CD experience
• Evidence of technical judgment — documented trade-off decisions
• Strong academic or technical track record (competitive programming, research, open-source) preferred
• Comfortable working with minimal structure and high ownership