Advanced Certificate in Generative Artificial Intelligence Apps Design and Prompt Engineering Module 1: Practical Approach to Large Language Models (LLMs) and Retrieval Augmented Generation (RAG)
About This Course
Since the advent of ChatGPT, Generative Artificial Intelligence (Gen AI) has emerged as a mainstream tool in the corporate world, transforming how organisations enhance productivity and learning, and in many cases, becoming a game changer. However, merely using ChatGPT or other pre-trained models is no longer sufficient as organisations shift from passive users to active creators and deployers of the technology. Moreover, effective governance and management of Gen AI are essential to ensure its responsible use. This Advanced Certificate is tailored for non-technical professionals, particularly those from small and medium-sized enterprises (SMEs), who have domain expertise but lack programming skills. It is designed for individuals interested in using prompt engineering techniques to develop Gen AI tools and chatbots for specific use cases, with the ability to deploy these applications either internally or to clients.
Module 1 provides an in-depth exploration of Large Language Models (LLMs) and Retrieval Augmented Generation (RAG), focusing on real-world applications, critical analysis, and practical insights. Participants will gain familiarity with popular LLMs, including OpenAI’s GPT, Anthropic’s Claude, Google’s Gemini, and Meta’s Llama, covering use cases, strengths and weaknesses, foundation models, and benchmarks across multi-modal applications. They will also understand the use cases for RAG and how this can be implemented using LLMs. Additionally, the module introduces Project Moonshot, an innovative, open-source toolkit for benchmarking, and baseline testing, designed to enhance the safety and security of LLMs. Through hands-on activities, participants will examine applications, implications, risk management, and internal implementation of Gen AI using custom datasets with minimal technical overhead. By the end of the module, participants will have developed, deployed, and learned to govern their own customised AI knowledge solution using RAG.
What You'll Learn
• Recognise the fundamental concepts of Large Language Models (LLMs) and Retrieval Augmented Generation (RAG), including stages, applications, and implications
• Explain how LLMs and RAG operate within an organisational context and identify their benefits and challenges
• Implement LLMs for specific use cases using Capabara Knowledge System (CKS), demonstrating RAG techniques to enhance knowledge system accuracy
• Examine risks associated with each of the 5 stages of LLMs and evaluate ways to mitigate these risks
• Assess LLM safety and security with Project Moonshot, incorporating feedback mechanisms for continuous improvement
• Apply feedback mechanisms for ongoing LLM improvements in CKS
• Develop custom knowledge solutions using Capabara's artificial intelligence tools, tailored to an organisation's specific needs
Entry Requirements
A tertiary education or diploma with at least three years of relevant work experience.