Site Reliability Engineer

Northeastern University•Boston, MA

16d

About The Position

The National Deep Inference Fabric (NDIF), an AI research infrastructure project funded by the National Science Foundation and private foundations, is seeking an experienced Site Reliability Engineer to improve reliability and coordinate day-to-day operations of its growing research platform. The role involves creating systems that keep distributed services healthy and available, improving cluster debugging and operations, and building tooling to support growth in demand and deployments. It is a hands-on position with significant latitude to assess complex technical problems, set priorities, and deliver innovations with minimal oversight. A research background is not required; knowledge of machine learning operations (MLOps) is highly desirable. The successful candidate will have production systems experience, a commitment to operating systems well, and motivation to contribute to AI safety and interpretability.

Requirements

Bachelor's degree in computer science, engineering, or a related field, or equivalent practical experience.
At least 3 years of experience in site reliability engineering (SRE) or a closely related role.
Experience operating distributed systems in production, with a track record of owning complex systems end to end.
Strong Python skills and comfort working with backend services.
Experience with observability tooling such as Prometheus and Grafana, or equivalent tools.
Working command of CI/CD pipelines, automated testing, and infrastructure as code.
Judgment to debug and optimize systems with minimal guidance, along with communication skills that keep teams aligned.
Genuine interest in AI safety, interpretability, and supporting interdisciplinary research.

Nice To Haves

Knowledge of MLOps.

Responsibilities

Measure and improve the reliability, availability, and performance of the platform used by research teams.
Build and maintain observability systems so issues are detected early and debugging is proactive.
Improve operational tooling and automation, including CI/CD, automated testing, and infrastructure as code.
Create effective workflows for debugging and resolving issues across the distributed stack, including Ray connectivity, GPU and resource allocation, and scheduling.
Partner with the broader NDIF team to keep the system dependable and usable as user needs evolve.
Establish and advocate for operational best practices as the platform grows.