ML Infrastructure Platform Engineer, Core Inference

Snap Inc.; Palo Alto (CA); On-site; USD 133,000 - 235,000; Full time

Benefits offered by this job

Paid parental leave

Comprehensive medical coverage

Mental health support

Job summary

Snap Inc. is seeking a Software Engineer for the Core Inference team within the ML Platform. You’ll work on the end-to-end inference stack powering Snap’s personalized experiences across Discover, Stories, and more.

You will optimize models for production, build scalable serving infrastructure, and collaborate with Ads, Content, Growth, and other teams to enable high-quality AI features at scale.

Qualifications

• Bachelor’s degree in a technical field or equivalent experience.
• 2+ years post-bachelor software development experience or advanced degree with relevant experience.
• Experience building large-scale production ML systems or distributed systems.

Responsibilities

• Support models powering personalized experiences with ranking needs.
• Optimize and prepare models for production use across training and inference.
• Own infrastructure to host and serve recommendation models and self-hosted LLMs.
• Work on LLM fine-tuning, reinforcement learning infrastructure, serving and inference optimization.
• Provide model-hosting and serving capabilities to multiple Snap product areas.

Skills

Python

Java

Distributed systems

Big data processing

Collaboration

Reliability at scale

Learning new concepts

AI systems & tools

Education

Bachelor’s degree in computer science or equivalent

Master’s degree or PhD in a technical field (advantage)

Tools

TensorFlow

PyTorch

Caffe2

Spark ML

scikit-learn

Job description

Snap Inc. is seeking a Software Engineer for the Core Inference team within the ML Platform. You’ll work on the end-to-end inference stack powering Snap’s personalized experiences across Discover, Stories, and more.

You will optimize models for production, build scalable serving infrastructure, and collaborate with Ads, Content, Growth, and other teams to enable high-quality AI features at scale.