ML Infra Engineer — Scale Compute for AI Research

Alpaca Research; Cambridge (MA); On-site; USD 150,000 - 210,000; Full time

Job summary

Alpaca Research is based in Cambridge, MA, and is seeking an ML infra engineer to oversee the training infrastructure for NLP and Robotics research teams. You will set up and scale compute resources, enforce efficient utilization, and simplify access for researchers to accelerate experiments.

You will collaborate with researchers to tailor infrastructure to workflows, manage permissions, and engage with vendors to optimize costs and performance.

Qualifications

• Familiar with AWS, GCP, Azure, Slurm, Kubernetes.
• Proven track record of building & maintaining cloud-based infrastructure.
• Familiar with distributed infra & multi-host setups.
• Strong communications skills and independence in owning projects.
• Familiar with building testing pipelines, CI/CD, logging/telemetry for ML infra use

Responsibilities

• Setting up and scaling our compute resources.
• Enforcing efficient compute utilization.
• Communicating with compute vendors to secure competitive rates.
• Making compute resources easy for researchers to use.
• Managing permissions & onboarding/offboarding users

Skills

Cloud infra

CI/CD pipelines

Distributed systems

Communication skills

Independence in owning projects

Tools

AWS

GCP

Azure

Slurm

Kubernetes

Job description

Alpaca Research is based in Cambridge, MA, and is seeking an ML infra engineer to oversee the training infrastructure for NLP and Robotics research teams. You will set up and scale compute resources, enforce efficient utilization, and simplify access for researchers to accelerate experiments.

You will collaborate with researchers to tailor infrastructure to workflows, manage permissions, and engage with vendors to optimize costs and performance.