Senior Remote Model Optimization Engineer

Bright Vision Technologies; Round Rock (TX); On-site; USD 150,000 - 175,000; Full time

Job summary

Bright Vision Technologies is seeking a Model Optimization Engineer to maximize throughput, reduce latency, and lower cost across training and inference for large neural networks. You will optimize from kernel code to distributed systems, collaborating with product, design, and operations teams.

The ideal candidate has strong Python and C++, experience with modern GPUs, and a track record shipping production AI workloads.

Qualifications

• Bachelor's or Master's degree in CS/CE or related field.
• 6+ years in performance engineering, ML systems, or HPC.
• Strong Python and C++ proficiency.
• Hands-on GPU optimization experience for DL workloads.
• Understanding of distributed training and inference.
• Experience with profiling CPU/GPU/distributed systems.
• Familiar with model compression and memory hierarchies.
• Excellent measurement, debugging and analytical skills.
• Strong communication and collaboration abilities.

Responsibilities

• Optimize neural network workloads from kernel to distributed system level.
• Reduce latency, increase throughput and lower costs for training/inference.
• Collaborate with product, design, engineering, and operations teams.
• Mentor junior engineers and contribute to code/design reviews.

Skills

Python

C++

Performance engineering

ML systems

GPU optimization

Distributed training

Profiling tools

Memory management

Communication

Team collaboration

Education

Bachelor's or Master's in Computer Science / Computer Engineering

Tools

Triton

CUTLASS

TensorRT

DeepSpeed

Job description

Bright Vision Technologies is seeking a Model Optimization Engineer to maximize throughput, reduce latency, and lower cost across training and inference for large neural networks. You will optimize from kernel code to distributed systems, collaborating with product, design, and operations teams.

The ideal candidate has strong Python and C++, experience with modern GPUs, and a track record shipping production AI workloads.