Model Performance Engineer

Job Summary

A company is looking for an AI Engineer - Model Performance to optimize model inference and build fine-tuning infrastructure.

Key Responsibilities

• Optimize inference performance for models, focusing on speed and cost-effectiveness
• Develop fine-tuning pipelines to streamline the model training process
• Debug production inference issues and evaluate serving frameworks for optimal performance

Required Qualifications

• Deep experience with LLM serving frameworks and tuning strategies
• Hands-on quantization experience with a strong understanding of various techniques
• Production fine-tuning experience with familiarity in training frameworks
• Strong proficiency in Python for infrastructure and pipeline development
• Comfort with GPU profiling and performance analysis