Associate Tech Specialist

AI/ML Model Deployment Specialist / Production MLOps Engineer

Location:

San Jose, CA

Job Description:

AI/ML Model Deployment Specialist / Production MLOps Engineer

Overview:
We are seeking a highly skilled and specialized AI/ML Model Deployment Specialist / Production MLOps Engineer. The core focus of this role is to take existing machine learning models and optimize their deployment and serving infrastructure for high-performance, production-ready inference, with a strong emphasis on leveraging state-of-the-art AI model serving technologies.

Responsibilities:

• Model-Specific Optimization:
  Analyze and understand the underlying logic and dependencies of various AI/ML models (primarily using PyTorch and Tensor Flow) to identify bottlenecks in the inference pipeline.
• High-Performance Serving Implementation:
  Design, implement, and manage high-performance inference serving solutions utilizing specialized inference servers (e.g., vLLM) to achieve low latency and high throughput.
• GPU Utilization Optimization:
  Optimize model serving configurations specifically for GPU hardware to maximize resource efficiency and performance metrics in a production environment.
• Containerization for Deployment:
  Create minimal, secure, and production-ready Docker images for streamlined deployment of optimized models and inference servers across various environments.
• Collaboration:
  
  Work closely with core engineering and data science teams to ensure a smooth transition from model development to high-scale production deployment.

Required Skillsets:

• AI/ML Domain Expertise:
  Deep understanding of the AI/ML domain, with the core effort centered around model performance and serving, rather than general infrastructure.
• ML Frameworks Expertise in PyTorch and Tensor Flow:
  Proven ability to work with and troubleshoot model-specific dependencies, logic, and graph structures within these major frameworks.
• Inference Optimization Production Inference
  
  Experience:
  
  Expertise in designing and implementing high-throughput, low-latency model serving solutions.
• Specialized Inference Servers:
  Mandatory experience with high-performance inference servers, specifically including vLLM, or similar dedicated LLM serving frameworks.
• GPU Optimization:
  Demonstrated ability to optimize model serving parameters and infrastructure to maximize performance on NVIDIA or equivalent GPU hardware.
• Deployment and Infrastructure Containerization (Docker):
  Proficiency in creating minimal, secure, and efficient Docker images for model and server deployment.
• Infrastructure Knowledge (Helpful, but Secondary):
  General knowledge of cloud platforms (AWS, GCP, Azure) and Kubernetes/orchestration is beneficial but the primary focus remains on model serving and optimization.

Qualifications:

• Bachelor's or Master's degree in Computer Science, Engineering, or a related field, or equivalent practical experience.
• 8+ years of experience in MLOps, Machine Learning Engineering, or a specialized Inference Optimization role.
• A portfolio or project experience demonstrating successful deployment of high-performance, containerized ML models to production scale.