Courses and applied labs
Study LLM foundations and practical applications through supervised implementation exercises.
Work produced Course notebooks, a working prototype and an assessment.
PROGRAM IN DEVELOPMENT
LLM
Learning Lab.
LLM principles, applications
and AI safety.
The program brings together technical courses, applied labs, academic seminars and mentored research. Topics cover how language models work, how to build applications with them, and how to evaluate their accuracy, cost and risks.
Discuss the programFor students, educators and research collaborators.
LLM PRINCIPLES & APPLICATIONS
Technical curriculum
Model foundations, application development and evaluation.
-
01
Model foundations
Tokens and representations; attention and Transformers; pre-training, post-training and inference.
-
02
Prompts and retrieval
Prompt design, retrieval-augmented generation, source selection and citation quality.
-
03
Adaptation and agents
Supervised adaptation, preference learning, tool use, permissions and structured workflows.
-
04
Evaluation and delivery
Test sets, reproducibility, error analysis, latency, cost and human review.
PROJECT DIRECTIONS
- An academic-reading assistant with citations.
- A cultural-heritage learning application with an evaluation report.
TEACHING & RESEARCH
How the program
is taught.
Taught exercises lead into guided replication and, when the student is ready, a scoped research investigation.
Seminars and academic projects
Read and discuss research with subject experts, join interdisciplinary reading groups and develop a project around the material.
Work produced A literature review, project brief and reviewed presentation.
Mentored research
Agree on a research question and evaluation method, reproduce a baseline, run controlled studies and discuss the findings with a mentor.
Work produced Reproducible code, an experiment log and a research report.
AI COMPLIANCE & SAFETY
Practical responsibilities
alongside technical work.
Each area connects a topic with a document or review that students can use in their own projects.
Data rights and privacy
Consent, permitted use, licensing, data minimization and handling sensitive student information.
Data-use and retention checklist
Academic integrity
Student authorship, attribution and clear disclosure of where and how AI was used.
Project integrity statement
Governance and evaluation
System limitations, bias, reliability, risk ownership, documentation and escalation.
System card and risk register
AI security
Prompt-injection awareness, tool permissions, secure code review and analysis of synthetic security logs.
Instructor-reviewed safety report
Security exercises use approved training assets and supervised practice.
LLM LEARNING LAB