Senior ML Platform Engineer - Scalable MLOps

Reddit, Inc.; San Francisco (CA); On-site; USD 217,000 - 303,000; Full time

Benefits offered by this job

Equity

Medical Insurance

Dental Insurance

Vision Insurance

401(k) employer match

Generous vacation

Parental leave

Job summary

Reddit, Inc. is seeking a Senior ML Infrastructure Engineer to lead the design of a platform for large-scale ML models.

You will own end-to-end lifecycle patterns, from data prep to model deployment, and drive efficiency across distributed training environments. You will collaborate with ML engineers to optimize performance, memory usage, and GPU costs while building scalable graph-based workflows and data pipelines using modern data tools.

Qualifications

• 5+ years of experience in ML infrastructure, including model training and deployments.
• Experience with memory and GPU profiling for ML workloads.
• Cloud-based platforms and IaC (e.g., Terraform) for ML platforms.
• Experience with MLflow or WandB for experiment tracking and model registries.
• Proficiency in Python and ML frameworks (PyTorch, TensorFlow).
• Experience with distributed training frameworks (Ray, Kubernetes).
• Familiarity with graph databases (Neo4j, JanusGraph, TigerGraph) and graph ML frameworks.

Responsibilities

• Design end-to-end ML lifecycle patterns (MLOps) to boost development velocity.
• Build and support a graph ML platform that scales models and workflows.
• Collaborate on performance tuning, training efficiency, and GPU cost reductions.
• Optimize batch data processing in a data warehouse using Beam, Spark, and Ray Data.
• Architect pipelines for billions of graph nodes and edges.

Skills

ML infrastructure

Python

PyTorch

TensorFlow

Kubernetes

GPU profiling

distributed training

graph ML

communication

Tools

GCP BigQuery

Google Cloud Storage

Terraform

MLflow

Weights & Biases (WandB)

PyTorch Geometric

Deep Graph Library

Neo4j

JanusGraph

TigerGraph

Job description

Reddit, Inc. is seeking a Senior ML Infrastructure Engineer to lead the design of a platform for large-scale ML models.

You will own end-to-end lifecycle patterns, from data prep to model deployment, and drive efficiency across distributed training environments. You will collaborate with ML engineers to optimize performance, memory usage, and GPU costs while building scalable graph-based workflows and data pipelines using modern data tools.