From December to February, you learn with OUTTA’s own curriculum and build things yourself.
ProgramYou build your foundations with the curriculum and complete at least three mini-projects. Everyone covers data analysis and applied AI; you also explore at least one of ML, CV or NLP.
You pick a problem you care about and carry it through implementation and analysis. The work is mostly remote, and you explain not only your method and results but also their limits.
It builds the groundwork for advanced study and practical AI skills.
An introduction to data science, data preprocessing and visualization.
How to keep up with fast-moving AI research. You work with Hugging Face and read recent work together.
How to use AI well, and what it still can’t do.
Machine learning and deep learning, with theory and practice in computer vision and natural language processing. Basic math and some experience writing Python are enough to follow.
Core theory and practice of machine learning and deep learning.
Starting from CNNs, you learn and implement segmentation, object detection and GANs.
From classic language models through RNNs and seq2seq to Transformers, BERT and GPT, with Python practice on the ideas behind today’s LLMs.
These are the fields you can choose for mini-projects. Research topics and team introductions are shared before you choose a team.
An introduction to Python and AI, written so middle-school students can get comfortable with programming and algorithms.
View on Kyobo Python Foundations for AI (opens in a new tab)
Machine-learning theory and Python practice that anyone with high-school math can follow.
View on Kyobo First Steps in Machine Learning (opens in a new tab)
For readers who know machine learning, a hands-on look at how computers learn language.
View on Kyobo First Steps in Natural Language Processing (opens in a new tab)I didn’t know the first thing about data, and I came away with a lot of concepts and knowledge. The team replied fast and kindly, even to messages sent at dawn, so I could ask anything without worrying.
We were encouraged to use ChatGPT on assignments, so I could focus on the real questions rather than the code. Working out the direction of our team project felt close to real industry work.
What helped most was how the mentors walked through what each part of the code meant and did, step by step. Being able to join fully online was great too.
In a short time I learned data management and LLMs through high-quality lectures and hands-on practice. It was the best thing I did all break.