AI Infrastructure Engineer - GPU & Real-Time Ad Systems

Mintegral; Seattle (WA); On-site; USD 140,000 - 210,000; Full time

Job summary

Mintegral is seeking exceptional Software Engineers who are passionate about AI infrastructure, high-performance computing, and building reliable systems at massive scale. This role targets new graduates and early-career engineers at the intersection of distributed systems and ML infrastructure.

Join a team building AI-powered ad decision systems, serving billions of requests with GPU-accelerated training and inference pipelines.

Qualifications

• Bachelor's, Master's, or Ph.D. in Computer Science, Software Engineering, AI, Data Science, or related fields.
• Strong programming skills in C++, Python, Java, Go, or similar languages.
• Strong foundation in algorithms, data structures, operating systems, distributed systems, and computer architecture.
• Strong interest in machine learning systems, AI infrastructure, and high‑performance computing.

Responsibilities

• Design and build high-performance AI infrastructure supporting large-scale advertising machine learning systems.
• Develop GPU-accelerated machine learning training and inference infrastructure, including distributed training and model optimization.
• Build and optimize ML systems using PyTorch, TorchRec, and distributed training frameworks.
• Design and implement high-throughput, low-latency real-time decision systems serving massive volumes of advertising requests.
• Build data and feature pipelines for embeddings, multimodal features, and real-time feature serving.
• Develop scalable model serving infrastructure for production ML applications.
• Improve system performance through parallel computing, GPU acceleration, and memory optimization.
• Collaborate with ML engineers and researchers to productionize state-of-the-art models.

Skills

C++

Python

Java

Go

Distributed systems

GPU programming

Machine learning systems

English communication

Education

Bachelor's degree in Computer Science

Master's degree in related field

Ph.D. in relevant field

Tools

PyTorch

TorchRec

DeepSpeed

CUDA

Distributed training frameworks

Job description

Mintegral is seeking exceptional Software Engineers who are passionate about AI infrastructure, high-performance computing, and building reliable systems at massive scale. This role targets new graduates and early-career engineers at the intersection of distributed systems and ML infrastructure.

Join a team building AI-powered ad decision systems, serving billions of requests with GPU-accelerated training and inference pipelines.