Tower Research Capital
Research Platform Engineer
$200K–$300K
New York +1 · Full-time · Posted Oct 1, 2026
- Location
- New York
New York, NY, United States
About the job
Develop a multi-tenant research compute platform that orchestrates large-scale machine learning workloads across hybrid GPU and CPU infrastructure.
What you'll do
- Build platform APIs, job control, and scheduling abstractions for research simulations, training, and pipelines.
- Implement multi-tenant scheduling, isolation, and execution portability across datacenters and public clouds.
- Improve pipeline reliability, data-path performance, observability, and compute cost attribution.
What they ask for
- Deep experience with HPC job schedulers, including scheduling and containerized HPC execution.
- Advanced understanding of Kubernetes architecture, HPC-focused operators, and GPU-native schedulers.
- Experience designing vendor-agnostic infrastructure, cloud bursting, and compute across datacenters and public clouds.
- Proven track record building cluster-wide telemetry and cost-attribution platforms for multi-tenant environments.
Pay
$200,000-$300,000 USD base salary
Tools
Slurm, Kubernetes, Ray, PyTorch, JAX
AI focus
Build and operate compute infrastructure for distributed machine learning workloads, including training pipelines.
Model-written summary · the original posting is authoritative