ML Infrastructure Engineer, Core Inference

SNAP, Inc.; Bellevue (WA); On-site; USD 157,000 - 235,000; Full time

Benefits offered by this job

RSUs

Paid parental leave

Medical coverage

Mental health programs

Job summary

Snap Inc. is seeking a Software Engineer for the Core Inference team within the Core ML Platform. You will help power personalized experiences by building and maintaining ML infrastructure for production inference at scale.

You will collaborate across training and deployment pipelines, optimize models for low latency, and support hosting of models across Ads, Content, Growth, Lenses, Security, and My AI. Equity in RSUs is available, and Snap emphasizes in-office collaboration.

Qualifications

• Bachelor’s degree in a technical field such as computer science or equivalent experience.
• 2+ years of post-Bachelor’s software development experience; or Master’s +1 year; or PhD in a relevant field.
• Experience building large-scale production machine learning systems or distributed systems.

Responsibilities

• Support models powering personalized experiences across Snap products.
• Optimize and prepare models for production across training and inference.
• Own infrastructure to host/serve recommendation and embedding-generation models and LLMs.
• Work on LLM fine-tuning, RL infrastructure, serving, and inference optimization.
• Provide hosting/serving capabilities to multiple teams across Ads, Content, Growth, etc.
• Own inference components powering advertising models and ranking strategies.

Skills

Python

Java

Distributed systems

Big data

Collaboration

Learning

Scalability

Inference

Education

B.S. in CS or related

M.S. or PhD (preferred)

Tools

Spark

Flink

Ray

TensorFlow

PyTorch

Caffe2

Spark ML

scikit-learn

Job description

Snap Inc. is seeking a Software Engineer for the Core Inference team within the Core ML Platform. You will help power personalized experiences by building and maintaining ML infrastructure for production inference at scale.

You will collaborate across training and deployment pipelines, optimize models for low latency, and support hosting of models across Ads, Content, Growth, Lenses, Security, and My AI. Equity in RSUs is available, and Snap emphasizes in-office collaboration.