Intern AI Inference Optimization Engineer

@ IBM

Introduction

IBM Research shapes the future of technology with groundbreaking research and practical innovations. As a part of IBM Research, you will work with top researchers and developers to bridge AI advancements and hybrid cloud solutions, creating lasting impact.

Your Role And Responsibilities

As an Intern AI Inference Optimization Engineer at IBM, you will contribute by integrating AI research into software solutions. Your responsibilities include constructing and optimizing the AI platform technology stack and implementing software components on specialized AI hardware.

Key Duties

• Apply AI model development and training techniques.
• Perform foundation model inference and deployment using containerized programming paradigms.
• Integrate innovative LLMs, including Hybrid Mixture of Experts models, using libraries and frameworks such as PyTorch, TensorFlow, vLLM, and Hugging Face Transformers, TRL.
• Enhance data handling and pre-processing techniques for NLP tasks using open source libraries.
• Design and execute performance evaluation and benchmarking using simulated and observed techniques.

Preferred Education

Bachelor's Degree

Required Technical And Professional Expertise

• Student enrolled in a Master's or Ph.D. program in Computer Science or related fields.
• Strong programming skills in Python, Java, or C/C++.
• Proficiency in scalable software engineering with a focus on AI or machine learning.
• Understanding of machine learning algorithms and model serving frameworks like vLLM, TensorFlow Serving, or TorchServe.
• Experience with ML frameworks such as TensorFlow, PyTorch, Keras, and Scikit-Learn.
• Proficiency in using version control systems like Git.
• Proven contributions to open-source AI projects.

Preferred Technical And Professional Experience

• Experience in training and validating machine learning models for NLP.
• Implementing and fine-tuning pre-trained models using Hugging Face Transformers.
• Expertise in containerization technologies such as Docker and orchestration platforms like Kubernetes.
• Ability to deploy AI models for low latency inference.
• Skills in hyperparameter tuning and model optimization.
• Experience with GraphQL and model compression techniques.

Key skills/competency

• AI
• Inference
• Optimization
• Containerization
• Hybrid Cloud
• Software Engineering
• Benchmarking
• Machine Learning
• NLP
• Open-Source