Principal Model Optimization Engineer

Achievements

Optimized machine learning models for performance on GPU architectures, focusing on both training and inference workflows.

Responsibilities

• Conducted low-level performance profiling analysis to identify bottlenecks in existing machine learning pipelines and proposed actionable improvements.
• Developed best practices and tooling for model optimization and deployment.
• Collaborated with cross-functional teams, including data scientists and software engineers, to integrate and deploy optimized models into production environments.
• Partnered across organizations to build tooling, interfaces, and visualizations that make the ML@Roblox a delight to use.

Requirements

• Significant experience debugging GPUs - reading GPU profiles, debugging Xid errors, etc.
• Proficient in advanced tools and frameworks (e.g., CUDA, Triton, TensorRT) to enhance model execution speed and reduce latency.
• Experience with model optimization techniques for LLMs, such as speculative decoding, continuous batching, quantization, etc.
• A performance nut; passionate about pushing the limits of what’s possible.
• A generalization advocate; passionate about building tools and frameworks that consistently deliver improvements in model performance.
• Passionate about supporting internal partners (data scientists and ML Engineers) to meet and understand their needs.

Qualifications

• Bachelor's degree in Computer Science, Computer Engineering, Data Science, or a similar technical field.

Benefits

Equal employment opportunities are provided to all employees and applicants for employment and prohibited discrimination and harassment of any type without regard to race, color, religion, age, sex, national origin, disability status, genetics, protected veteran status, sexual orientation, gender identity or expression, or any other characteristic protected by federal, state or local laws.

Pay

The starting base pay for this position is as shown below. The actual base pay is dependent upon a variety of job-related factors such as professional background, training, work experience, location, business needs and market demand. Therefore, in some circumstances, the actual salary could fall outside of this expected range. This pay range is subject to change and may be modified in the future.

Schedule

Roles that are based in an office are onsite Tuesday, Wednesday, and Thursday, with optional presence on Monday and Friday (unless otherwise noted).