Staff AI Inference Engineer — Multi-Cloud, Low-Latency GPUs

USA023 Thomson Reuters Applications Inc; Eagan (MN); Hybrid; USD 127,000 - 237,000; Full time

Benefits offered by this job

Hybrid Work Model

Mental Health Days off

Headspace app access

Tuition reimbursement

Employee Stock Purchase Plan

Job summary

Thomson Reuters is seeking a Staff Inference Engineer, AI, to productionize and scale AI/LLM workloads across AWS, Azure, GCP, and OCI. You will collaborate with Product, Data Science, and Platform Engineering to onboard new research models into production and optimize latency on GPU hardware.

The role requires 7+ years of experience, strong Python and C++ skills, and hands-on GPU programming with CUDA, TensorRT, and ONNX Runtime. Expect hybrid work in the United States.

Qualifications

• 7+ years of relevant experience.
• Proficiency in Python and at least one systems language (C++ preferred).
• Strong understanding of ML/LLM fundamentals and inference optimization techniques.
• Experience deploying AI workloads to AWS/GCP/Azure and Kubernetes.

Responsibilities

• Productionize, optimize, and scale AI/LLM workloads powering TR products.
• Run models efficiently across multi-cloud footprint with enterprise reliability.
• Build and optimize containerized inference pipelines using Kubernetes.
• Implement routing and failover for AI traffic and integrate models into production APIs.
• Collaborate with platform teams to align inference workloads with cloud-native patterns.

Skills

ML/LLM fundamentals

Python

C++

GPU programming

Cloud & Kubernetes

Tools

CUDA

TensorRT

ONNX Runtime

PyTorch

TensorFlow

Kubernetes

AWS

GCP

Azure

Job description

Thomson Reuters is seeking a Staff Inference Engineer, AI, to productionize and scale AI/LLM workloads across AWS, Azure, GCP, and OCI. You will collaborate with Product, Data Science, and Platform Engineering to onboard new research models into production and optimize latency on GPU hardware.

The role requires 7+ years of experience, strong Python and C++ skills, and hands-on GPU programming with CUDA, TensorRT, and ONNX Runtime. Expect hybrid work in the United States.