Lead/Architect – GPU Programming & Infrastructure - LAGPU

Position Title: Lead/Architect – GPU Programming & Infrastructure
Location: Bronx, NY
Contract Duration: 6 Months
Employment Type: Contract

Position Overview:

The Lead/Architect will play a critical role in deploying and optimizing LLMs (e.g., LLaMA), managing GPU-based inference infrastructure, and streamlining production-grade machine learning pipelines. This is a highly collaborative role requiring strong engineering skills, cloud expertise, and a deep understanding of model performance tuning.

Key Responsibilities:

• Deploy and optimize LLMs using Hugging Face Transformers

• Configure and manage GPU-based inference workflows with NVIDIA CUDA

• Automate infrastructure and development tasks using Python, PyTorch, and shell scripting

• Administer and maintain Linux (Ubuntu) environments for development and deployment

• Provision and manage GPU resources in cloud environments (AWS EC2, GCP, Hugging Face Spaces)

• Implement robust environment isolation and package management using Conda and pip

• Lead the end-to-end design and deployment of scalable ML model pipelines

• Partner with cross-functional engineering teams to ensure optimal model performance and reliability

Qualifications:

• Strong experience in deploying and fine-tuning LLMs in production environments

• Proficiency with Hugging Face Transformers and PyTorch

• Deep knowledge of GPU performance tuning and parallel processing

• Fluency in Python and experience with Linux systems administration

• Hands-on expertise with cloud GPU orchestration and containerized environments

• Proven leadership in designing scalable AI/ML infrastructure

• Ability to work in a fast-paced, performance-driven environment