Instructor
Dr. Dong Li, Computer Science and Engineering, UC Merced
Imagine a summer where you build AI that feels straight out of your favorite movie or
video game: design a script-writing assistant that helps you classify your movies based
on your preferences, train an NPC that learns from player choices, and create an image-
to-concept generator that turns your fan art into polished poster art. You’ll work in small
teams on hands-on labs led by mentors who turn confusing theory into tools you can use,
iterate on portfolio-ready demos, and compete in a final capstone project that rewards
creativity, storytelling, and technical skill. Whether you’re into AI model design, or just
love making things that wow people, this program gives you real projects to show
colleges and employers, new friends who share your passions, and the confidence to
keep building.
Prerequisite
Students do not need advanced knowledge to join, but a few foundational skills and a
short preparatory plan will make the four-week project productive and enjoyable. The
prerequisites below are lightweight and geared to bring motivated high-schoolers up to
speed in about 1–2 weeks of guided prep.
Required Knowledge and Skills
Basic Python: variables, lists, loops, functions, and reading/writing files.
Comfort with computers: installing software, using a terminal or command
prompt, and managing files.
High school math: algebra and basic probability/statistics (mean, variance, basic
graphs).
Growth mindset: willingness to debug, try small experiments, and learn from
mistakes.
Strongly Recommended but Not Mandatory
Familiarity with Jupyter Notebooks or Google Colab for interactive coding.
Basic data literacy: reading CSVs, simple plotting.
Version control basics: very simple GitHub usage for pushing code (optional for
beginners)
General Description
This four-week project introduces high school students to the intersection of High
Performance Computing (HPC) and Artificial Intelligence (AI) through hands-on
experiments, team work, and a final capstone. Students will learn why large-scale
compute matters, how modern AI models are developed, and practical techniques to
speed up data processing and model training using parallel and distributed approaches.
The course balances conceptual framing with guided coding exercises and a real-world
project that students present at the end.
What students will do each week
Week 1 Foundations — Set up a Python environment, explore HPC concepts (CPUs,
GPUs, clusters), and run small benchmarks to compare CPU vs GPU performance.
Week 2 Machine Learning Essentials — Build and train simple models
(classification/regression), use libraries such as NumPy, Pandas, scikit-learn and a
deep learning framework, and visualize model behavior.
Week 3 Scaling and Parallelism — Learn multiprocessing, multithreading, and
basics of distributed training; experiment with data-parallel strategies and
simulate MPI-style workflows.
Week 4 Capstone Project — Apply skills to a chosen project (e.g., image classifier at
scale, recommendation system, or parallelized data analysis), produce
reproducible code, and deliver a short presentation/demo.
Key Learning Outcomes
Technical concepts: Understand HPC building blocks, GPU acceleration,
parallelism, and distributed computing patterns.
AI fundamentals: Grasp supervised learning, neural network basics,
training/validation workflows, and model evaluation.
Practical skills: Use Python for data handling and modeling; run benchmarks;
implement multiprocessing and simple distributed training.
Project practices: Collaborate with peers, manage code in a repository, document
experiments, and present results clearly.
Estimated Resources and Platforms
Personal laptops; optional access to cloud notebooks (Google Colab) for GPU time.
Recommended toolset: Python, Jupyter Notebooks, NumPy, Pandas, scikit-learn,
TensorFlow or PyTorch, and GitHub for code sharing.