image

AI Development Internship

HYAB Youth Start-up Internship Programme (YSIP) 2025

Company Introduction

We’re seeking a motivated university student with a strong foundation in deep learning and hands-on experience with PyTorch and Hugging Face. You’ll collaborate with our AI team to design, train, and deploy models while gaining mentorship from industry experts. This role is perfect for someone eager to bridge academic knowledge with practical AI development.

Job Function

Technology Research and Development

Job Description

  • Develop, optimize, and debug deep learning models using PyTorch.
  • Implement and fine-tune pre-trained models from Hugging Face (Transformers, Diffusers, etc.) for tasks like NLP, computer vision, or generative AI.
  • Collaborate with the team to integrate AI solutions into real-world applications.
  • Stay updated on advancements in AI/ML and propose innovative approaches.
  • Document code, experiments, and results for reproducibility.
  • Present findings and contribute to team discussions.

Work Mode

Part-time, Mix-mode (remain consistent within a calendar month)

Work Location

Science Park (Pak Shek Kok)

Preferred Academic Disciplines

Business & Management, Technology

Preferred Skills and Knowledge

Must-Have:
  • Pursuing a Bachelor’s/Master’s in Computer Science, AI, Data Science, or a related field.
  • Solid understanding of deep learning fundamentals (CNNs, RNNs, Transformers, etc.).
  • Proficiency in PyTorch (e.g., building custom models, training loops, tensor operations).
  • Familiarity with Hugging Face libraries (Transformers, Datasets, or Model Hub).
  • Strong coding skills in Python and experience with ML workflows (data preprocessing, evaluation metrics).
  • Problem-solving mindset and eagerness to learn.

Nice-to-Have:
  • Projects on GitHub showcasing PyTorch/Hugging Face implementations.
  • Exposure to NLP, CV, or generative AI (e.g., LLMs, diffusion models).
  • Basic knowledge of cloud platforms (AWS, GCP) or MLOps tools.
  • Understanding of CI/CD pipelines or model deployment.