Wintour

Artificial intelligence is profoundly transforming various industries, especially modern service sectors with close human interaction, such as hospitality, tourism, retail, food and beverage, and customer service. The application of Generative AI (GenAI) in particular shows immense potential within these areas. However, a common challenge arises when educational institutions conduct AI teaching: how can students move beyond just having "heard of" AI to truly gaining hands-on experience and mastering the practical skills needed to solve real-world problems in service scenarios using AI?

To help educational programs in various service industries keep pace with the changes of the digital age, we have developed a series of practical experimental courses on AI application development and practice. These courses progress from basic to advanced levels through seven core experimental tasks that are closely aligned with industry realities, highly hands-on, and progressively challenging. They are designed to help students systematically build an AI application mindset and acquire portable, practical operational skills.

• Systematic Progression, Spiral Advancement: Starting with foundational understanding of large models and prompt engineering, the curriculum gradually delves into workflow construction, RAG knowledge base applications, and complex dialogue flow design. It culminates in the design and analysis of intelligent agents (Agents) capable of autonomous decision-making and tool utilization. The content is arranged in a step-by-step manner, ensuring students progressively master core concepts and skills.

• Industry Frontier Alignment, Scenario-Driven Teaching: All experimental tasks are designed based on real-world service industry scenarios such as hotels, tourist attractions, and study tours. They cover typical applications including hotel financial analysis, customer service chatbots, scenic route planning, marketing, and study tour agents. This approach allows students to learn relevant technologies by solving practical business problems and understand the application value and challenges of GenAI.

• Emphasis on Practice and Analysis, Beyond Mere "Usage": While guiding students to interact with GenAI systems, the curriculum also encourages them to analyze the underlying working principles. Examples include the intent understanding of customer service chatbots, the retrieval mechanisms of RAG knowledge bases, the query transformation in NL2MQL2SQL*, and the decision-making and tool-calling mechanisms of intelligent agents. This fosters students' critical thinking skills to understand the essence beyond superficial phenomena.

• Comprehensive Coverage of Key AI Technical Points: This experimental series covers indispensable core technologies in current GenAI applications:

  • Large Language Models (LLM) - Fundamentals and Characteristics: The core technology.
  • The Key to Precise Intent: Prompt Engineering and Optimization: Crafting effective prompts.
  • The Bridge Connecting AI to the External World: GenAI Workflows and Patterns (RAG, NL2MQL2SQL): Structuring AI interactions and data access.
  • Giving AI Knowledge: Knowledge Base Construction and Management (Chunking, Vector Database Basics): Building and managing information for AI.
  • Enabling Intelligent Conversation: Chatflow, Intent Recognition, Multi-turn Dialogue, and Memory Management: Designing and managing conversational AI.
  • Enabling AI to Take Action: Agent Concepts, Tool Calling, and Agent Behavior Analysis: Designing and understanding autonomous AI.

• AI-Assisted Evaluation, Addressing Teaching and Evaluation Pain Points: The course employs an AI-assisted evaluation mechanism to solve the challenges of large teacher workloads and difficulty in providing personalized feedback to every student in traditional teaching models. By intelligently analyzing students' experimental work and reports, it provides instant, objective, multi-dimensional evaluations and specific improvement suggestions, helping teachers gain a more accurate understanding of student learning and supporting students' personalized development and skill enhancement.

• Ready-to-Use Teaching and Experimental Resources: Each experimental task is accompanied by detailed teaching objectives, a complete experimental process design, and detailed technical guidelines for technical personnel to set up the experimental environment. This greatly facilitates teacher preparation and experimental environment setup.

• Focus on Capability Cultivation and Quality Improvement: The entire experimental design incorporates objectives for cultivating higher-order skills such as scientific inquiry, critical thinking, systems analysis, problem-solving, and responsible AI application and ethical awareness, helping students become well-rounded future talents.

• AI Application Development and Practice System

Here is the English translation of the provided Chinese content, formatted for clarity:

1. LLM

   • Basic Knowledge: Learn about Large Language Models (LLM) and Prompts.
   • Advanced Knowledge: Understand important techniques for enhancing large models' reasoning abilities, such as Chain of Thought (CoT) and Reasoning-Action (ReAct).
   • Experimental Task 1: Initial Exploration of Large Models and Prompt Engineering.

2. Workflows

   • Basic Knowledge: Workflow + RAG+ Prompt.
   • Advanced Knowledge: Techniques for Chunking RAG Knowledge Bases.
   • Experimental Task 2: GenAI-Based Data Analysis Workflow.
   • Experimental Task 3: Planning and Implementation of a Hotel Operations Data Analysis Workflow.

3. Chatflows

   • Basic Knowledge: Chatflow + RAG+ Prompt.
   • Advanced Knowledge: Learn Intent Recognition and Large Model Memory Management.
   • Experimental Task 4: Planning and Implementation of a Hotel Intelligent Customer Service Chatbot.
   • Experimental Task 5: Planning and Implementation of a Scenic Area Intelligent Customer Service Chatbot.

4. Agents

   • Basic Knowledge: Orchestration and Design of Agents.
   • Advanced Knowledge: Tool Calling within Agents.
   • Experimental Task 6: Planning and Implementation of a Hotel Intelligent Marketing Agent.
   • Experimental Task 7: Planning and Implementation of a Travel Agency Intelligent Marketing Agent.