ML Network Development Engineer: Scale AWS Data Fabric

Socket.dev; New York (NY); On-site; USD 156,000 - 211,000; Full time

Job summary

AWS, based in New York, seeks a Network Development Engineer to join the Network Fabric Engineering team. You will build and operate networks to support ML training and inference workloads, balancing performance with reliability at scale.

Responsibilities include implementing changes, leading projects, and collaborating with automation teams to scale tooling. On-call work and data center travel are expected as part of a 24x7 environment.

Qualifications

• 4+ years of major internet routing protocols experience.
• 4+ years of working in a Linux/Unix environment experience.
• 4+ years designing large scale networks.
• 4+ years of operating large scale networks and operational excellence.
• 1+ years of automation scripting using Python, Bash, Shell and/or Perl.
• Knowledge of Machine Learning and LLM fundamentals, including transformer architecture, training/inference lifecycles, and optimization technique
• 4+ years of optimizing network performance with solid experience in the end-to-end stack (network and servers)

Responsibilities

• Create and implement changes on the network.
• Lead network projects in sustaining engineering, deployment/implementation, scaling, and optimization.
• Operate and troubleshoot routing, interconnectivity, platforms, performance, and configurations.
• Collaborate with automation teams to scale tooling.
• Maintain standards across the networking toolset and ensure compliance.

Skills

Networking

Linux/Unix

Network design

Network operations

Tools

Python

Bash

Shell

Perl

Job description

AWS, based in New York, seeks a Network Development Engineer to join the Network Fabric Engineering team. You will build and operate networks to support ML training and inference workloads, balancing performance with reliability at scale.

Responsibilities include implementing changes, leading projects, and collaborating with automation teams to scale tooling. On-call work and data center travel are expected as part of a 24x7 environment.