Production Systems Engineering Intern

Meta; Sunnyvale (CA); On-site; USD 34,000 - 44,000; Full time

Job summary

Meta, in its HDRTP group within the Infrastructure organization, is offering a 12-week summer internship in Menlo Park, CA. You will explore how Meta's servers are designed, learn provisioning and health-check systems, and work on automating debugging and anomaly-detection pipelines.

You will own a project from design to rollout and present outcomes to the team. Qualifications include enrollment in a related computer science/engineering degree, Python programming experience, and familiarity with

Qualifications

• Currently enrolled in a full-time degree program in Computer Science, Computer Engineering, Electrical Engineering or a related field, with an expected graduation date after the internship.
• Experience programming in Python.
• Experience working with SQL or Python data tools.

Responsibilities

• Learn how Meta's servers are designed and how components connect.
• Familiarize with internal systems to provision servers, run health checks, and remediate faults.
• Investigate failures on production servers and trace to root cause.
• Build software to automate debugging, diagnostics, or anomaly-detection pipelines.
• Own your project from design to rollout and present results to the team.

Skills

Python

Distributed systems

Linux

Data analysis

Education

Bachelor's/Master's in CS/CE/EE

Tools

SQL

Job description

We're part of Hardware Design and Release to Production (HDRTP) in Meta's Infrastructure organization. Meta designs its own servers, and our team supports them for the 5+ years they run in Meta's data centers. That includes the compute, database and storage servers that run Facebook, Instagram, WhatsApp and Messenger, along with the GPU and AI accelerator systems that train and serve Meta's Muse models. When these servers fail, we find out why and land durable fixes. Engineers on the team specialize in one of several areas: in-depth hardware root cause analysis, automation that makes debugging and triage faster, hardware diagnostics, or datasets and pipelines that detect anomalies earlier. Your internship project will focus on one of them.

This is a 12-week internship based in Menlo Park, CA.

Responsibilities

• Learn how Meta's servers are designed, including what each component does and how they connect to each other.
• Get familiar with the internal systems Meta uses to provision servers, run health checks and remediate faults.
• Investigate failures on production servers and trace them to a root cause.
• Build software that helps the team find and fix hardware failures faster, such as triage automation, hardware diagnostics or anomaly-detection pipelines.
• Own your project from design to rollout: work with the engineers who will use it, document it, and present your results to the team.

Qualifications

• Currently enrolled in a full-time degree program in Computer Science, Computer Engineering, Electrical Engineering or a related field, with an expected graduation date after the internship
• Programming experience in Python
• Coursework or project experience in distributed systems, computer networks or computer architecture
• Experience working in a Linux environment
• Must be available to work onsite in Menlo Park for a 12-week internship in summer 2027
• Must obtain work authorization in country of employment at the time of hire, and maintain ongoing work authorization during employment
• Intent to return to full-time degree program after completion of the internship
• Experience debugging hardware or low-level software, such as firmware, drivers or the Linux kernel
• Familiarity with server architecture: CPUs, GPUs, memory, storage and the interconnects between them, like PCIe or NVLink
• Experience building tools that automate debugging, monitoring or operations work, including personal or class projects
• Experience analyzing large datasets or system telemetry with SQL or Python data tools