Research Platform Engineer — Hybrid ML/HPC & Multi-Tenant

Tower Research Capital; New York (NY); Presencial; USD 200 000 - 300 000; Tempo integral

Vantagens oferecidas por esta oferta de emprego

Generous paid time off policies

Savings plans and other financial welI

Hybrid working opportunities

Free breakfast, lunch, and snacks

In-office wellness experiences

Company-sponsored sports teams and фит

Volunteer opportunities and charitable

Social events and learning

Workshops and continuous learning

Resumo da oferta

Tower Research Capital is a leading quantitative trading firm with a high-performance platform and independent trading teams. Tower has a 25+ year track record of innovation and a global footprint.

As part of Tower Research's Core Engineering team, you will develop a multi-tenant research compute platform capable of dynamically orchestrating large scale ML workloads across a hybrid compute infrastructure of GPUs and CPUs.

Qualificações

• Deep experience with HPC job schedulers like Slurm and related concepts.
• Advanced understanding of Kubernetes architecture and GPU-native schedulers.
• Experience with distributed ML frameworks (Ray) and DL frameworks (PyTorch/JAX).
• Hands-on experience evaluating and operating job graph orchestration frameworks.

Responsabilidades

• Elevate Researchers' Experience with platform abstractions and durable job control.
• Architect multi-tenant scheduling and isolation across trading desks.
• Enable cross-datacenter and cloud portability for compute workloads across on-prem and cloud.
• Standardize workflow orchestration for multi-stage ML pipelines.
• Build fault-tolerant pipelines with checkpointing and retries.
• Optimize data paths with high-speed storage and networks.
• Deliver telemetry and cost attribution for resource usage.

Conhecimentos

HPC Administration

Kubernetes-native Engineering

Distributed ML Computing

Workflow Orchestration

Multi-Platform Architecture

Storage & Network Performance

Compute Observability

Architectural Leadership

Ferramentas

Slurm

Kubernetes

Ray

PyTorch

JAX

Descrição da oferta de emprego

Tower Research Capital is a leading quantitative trading firm with a high-performance platform and independent trading teams. Tower has a 25+ year track record of innovation and a global footprint.

As part of Tower Research's Core Engineering team, you will develop a multi-tenant research compute platform capable of dynamically orchestrating large scale ML workloads across a hybrid compute infrastructure of GPUs and CPUs.