This course trains students to design, implement, and deploy LLM-based agentic solutions in business settings. Students will learn core concepts and common practices behind modern LLMs, including training, finetuning, prompting, retrieval augmentation and safety guardrails. The course will also introduce how LLM-based agentic AI can be deployed and integrated into existing data analytics, decision support and business processes. Through hands-on exercises in Python, students will gain practical experience using LLM APIs and prompt templates, finetuning lightweight LLMs, orchestrating LLMs agents, and experimenting with simplified RAG (Retrieval Augment Generation) pipelines. The course will also introduce key issues in evaluation, governance, and risk management, enabling students to critically assess both the potential and limitations of LLM agentic systems. For SBM students only.
Unlock more information by signing in with your HKUST account.