Machine Learning/AI Infrastructure Engineering Intern (AI Platform) PhD, Winter 2027

Netflix•Los Gatos, CA

6d•$40 - $85•Remote

About The Position

The AI Platform team builds the infrastructure that Netflix's ML and AI systems run on, from large-scale training platforms and post-training/offline infrastructure to GPU-optimized inference and serving. You'll work on infrastructure that's closely co-designed with modeling teams (model–system codesign), so this role suits PhD researchers who enjoy working at the intersection of systems and ML rather than pure modeling.

Requirements

Currently enrolled student pursuing a PhD in Computer Science, Distributed Systems, Systems, Networking, Machine Learning, Computer Engineering, or a related field
Research or applied experience in one or more of the following: Distributed systems, distributed training/serving infrastructure; ML training platforms, post-training or offline infrastructure; Inference and serving optimization, GPU-optimized inference; Model–system codesign
Proficiency in Python
Experience with systems languages (Go, C++, or Rust) is a strong plus
Familiarity with distributed compute frameworks (e.g., Ray, Kubernetes, Spark) and ML training/serving stacks
Curious, self-motivated, and excited about solving open-ended infrastructure challenges at Netflix scale
Strong written and verbal communication skills
Must be returning to school for at least one semester/quarter after the internship
Resume or CV with complete contact information (email, phone, mailing address) and a list of relevant coursework and publications (if applicable)
Short statement describing research experiences and interests, and (optionally) their relevance to Netflix Research

Nice To Haves

Publications or strong research alignment with systems-track venues (OSDI, SOSP, NSDI) or applied ML venues
Prior industry or internship experience in ML infrastructure

Responsibilities

Work on infrastructure that's closely co-designed with modeling teams (model–system codesign)
Solve open-ended infrastructure challenges at Netflix scale