Remote AI Optimization Engineer — Throughput, Latency & Cost

Bright Vision Technologies; Hillsboro (OR); Remote; USD 90,000 - 110,000; Full time

Job summary

Bright Vision Technologies is seeking an AI Optimization Engineer to optimize throughput, reduce latency, and lower costs across training and inference workloads for large neural networks. This role spans from low-level kernel tuning to distributed system optimization, with exposure to GPU architectures and model parallelism.

The successful candidate will collaborate with product, design, engineering, and operations to translate ambiguous requirements into robust, measurable solutions and will

Qualifications

• Bachelor’s or Master’s degree in CS, CE, or a related field.
• Six+ years of experience in performance engineering, ML systems, or HPC.
• Strong proficiency in Python and C++.
• Hands-on experience optimizing deep learning workloads on modern GPUs.
• Knowledge of distributed training and inference techniques.
• Experience with profiling tools across CPU, GPU, and distributed systems.
• Familiarity with model compression and its impact on accuracy.
• Solid understanding of memory hierarchies and parallelism strategies.
• Strong debugging, measurement, and analytical skills.
• Excellent communication and collaboration abilities.

Skills

Performance engineering

ML systems

HPC

Python

C++

GPU optimization

Distributed training

Inference techniques

Profiling tools

Model compression

Memory architecture

Debugging

Analytics

Communication

Education

Bachelor’s or Master’s degree in CS/CE or related field

Job description

Bright Vision Technologies is seeking an AI Optimization Engineer to optimize throughput, reduce latency, and lower costs across training and inference workloads for large neural networks. This role spans from low-level kernel tuning to distributed system optimization, with exposure to GPU architectures and model parallelism.

The successful candidate will collaborate with product, design, engineering, and operations to translate ambiguous requirements into robust, measurable solutions and will