AI Frameworks Engineer (openvino, Genai)

Intel Intel · Semiconductors · Leixlip, Ireland

AI Frameworks Engineer focused on optimizing generative AI models for efficient inference using the OpenVINO toolkit across Intel hardware, from edge to cloud. This role involves deep dives into generative model architectures and the OpenVINO ecosystem, implementing new features, and optimizing performance for state-of-the-art inference solutions.

What you'd actually do

  1. Spearhead the development and enhancement of innovative use cases for OpenVINO GenAI.
  2. Dive deep into the architecture of generative models and the OpenVINO ecosystem.
  3. Implement new features, and optimize performance to deliver state-of-the-art inference solutions.
  4. Empower developers worldwide by optimizing deep learning models for maximum performance across diverse Intel hardware.

Skills

Required

  • Bachelor's, Master's Computer Science, Computer Engineering, Mathematics, or a related field
  • Deep Learning architectures, particularly Generative AI models (e.g., Transformers, Diffusion models)
  • PyTorch, TensorFlow, and the Hugging Face Transformers library
  • C/C++, Python
  • multithreading concepts and practical experience in developing the thread-safe code
  • Strategic problem-solver
  • Impactful communicator and cross-functional collaborator
  • Effective verbal and written communication skills in English

Nice to have

  • model optimization techniques (i.e., quantization, pruning)
  • OpenVINO Toolkit or other deep learning inference runtimes(ONNX Runtime, Lite RT, Executorch)
  • contributing to open-source projects

What the JD emphasized

  • minimum of 5 years of hands-on development experience utilizing C/C++, Python
  • Proven experience in Deep Learning architectures, particularly Generative AI models
  • Familiarity with popular AI frameworks such as PyTorch, TensorFlow, and the Hugging Face Transformers library

Other signals

  • optimizing deep learning models for maximum performance
  • efficient and powerful generative AI pipeline deployments
  • implementing new features, and optimizing performance to deliver state-of-the-art inference solutions