AI Framework Software Intern

Intel Intel · Semiconductors · Shanghai, China

Internship role focused on optimizing AI software solutions, including algorithms, frameworks, and architectures for computer vision, machine learning, and deep learning. Responsibilities include researching model quantization and graph transformation, evaluating LLM performance on Intel platforms, analyzing software bottlenecks, and assisting in implementing and tuning AI models for performance and accuracy. The role emphasizes hardware-software integration and collaboration for scalable AI solutions.

What you'd actually do

  1. Conduct research and validation on emerging AI technologies, such as model quantization and graph structure transformation, deploying and optimizing proof-of-concept projects on Intel platforms.
  2. Perform performance evaluations on large language models (LLMs) on Intel platforms, developing functional modules for testing tools and ensuring effective benchmarking.
  3. Analyze performance test results to identify software bottlenecks, such as operator latency and memory usage, and propose optimization strategies to enhance inference speed and resource efficiency.
  4. Assist in implementing and tuning AI models for performance and accuracy, ensuring seamless hardware-software integration.
  5. Collaborate with internal product teams and external partners to create scalable AI software solutions that align with Intel's business goals.

Skills

Required

  • Master's degree in Computer Science, Computer Engineering, Software Engineering, or a related field.
  • Proficiency in at least one programming language, such as Python or C/C++.
  • Experience with widely used AI frameworks, including PyTorch and Hugging Face Transformers.
  • Hands-on experience with large language models and generative AI, with a demonstrated ability to learn and adapt quickly.

Nice to have

  • Strong problem-solving abilities and collaborative mindset, with experience analyzing and optimizing software performance.
  • A proactive approach to learning emerging AI technologies and contributing to innovative research.

What the JD emphasized

  • optimizing proof-of-concept projects on Intel platforms
  • performance evaluations on large language models (LLMs)
  • optimize inference speed and resource efficiency
  • implementing and tuning AI models for performance and accuracy

Other signals

  • optimizing AI software solutions
  • performance evaluations on large language models (LLMs)
  • optimize inference speed and resource efficiency
  • implementing and tuning AI models