- Introduces the course logic and public-safe boundaries of the site.
- Connects the GenAI landscape to the semester project format.
Academic Roadmap.
A public-facing deep dive from theoretical foundations to industrial implementation, bridging academic rigor and generative frontiers.
Topic 01
Introduction to GenAI and course structure
An editorial opening that frames generative AI, the semester arc, and the project-centered teaching method.
- Language-model fundamentals and token prediction.
- Transformer evolution and user-facing chat systems.
Topic 02
Large Language Models
Students move from language-model basics to the transformer era and study what is the attention, MOE, tokenization, RLHF for training the model.
Industrial insight
Guest lecture: AI agents
The course pivots toward tool use, coding, computer use agents, and multi-agent systems alongside an external industry perspective.
Topic 04
Prompt & Context engineering
The essential guideline for crafting the input for LLM to share the intent.
- Instruction following by the model
- Reasoning workflows for structured technical work.
- Multimodal model design and applications.
- Vision-language interaction and generation.
Topic 05
Diffusion models and vision
Students examine multimodal systems, generative imaging, and the shift from text-only interfaces to richer model interaction.
Checkpoint
Mid-semester project review
One week is reserved for prototyping, tutor feedback, and aligning project scope with the final delivery phase.
Core course philosophy
"We do not only teach how to prompt; we teach how to design systems around models so students can reason about utility, safety, and deployment together."
Topic 06
Practical deployment
The closing phase covers production-minded agentic systems, governance, open models, and the security & privacy.
- Attack and defence for responsible AI
- Open-source local AI model with license
Capstone achievement
Final presentation
- Live demo or prototype walkthrough
- Technical architecture documentation
- Testing, safety, and reflection
Finale
Final presentation
Students showcase their end-to-end generative AI systems through demos, technical framing, and deployment reflection.