Course Overview
Equip participants with the knowledge and skills to design, build, and deploy agentic AI systems for customer interactions.
Who should attend
- Application Developers
- Data Scientists
- Project Managers
- System Engineers
- Technology Trainers
Prerequisites
- Basic understanding of AI, familiarity with Python, and exposure to generative AI concepts (e.g., LLMs)
Course Objectives
- Understand what agentic AI is, how it differs from generative AI, and its role in customer interactions
- Building and deploying an autonomous AI Agent
- Dive into the technical components and start building a simple agentic AI
- Focus on autonomy, planning, and scalability in agentic AI systems
- Explore practical applications and learn to deploy agentic AI in real-world settings
- Address ethical challenges, evaluate performance, and showcase projects
Outline: Agentic AI – Building Autonomous Customer Interaction Systems (AA-BACIS)
Day One: Foundations of Agentic AI
1. Introduction to Agentic AI
- Definition and core chrematistics
- Autonomy, reasoning, action-oriented
- Evolution from generative AI to Agentic AI
- Key use cases in customer-facing roles (i.e. support, sales, engagement)
- Interactive Q&A: What makes AI “Agentic”
2. Technology Stack Overview
- Large Language Models (LLMs) for reasoning and communication
- Retrieval-Augmented Generation (RAG) for real-time access
- APIs and integrations for action-taking
- Memory and context management systems
- Demo: Compare a generative chatbot vs. an agentic AI workflow
3. Customer Interaction Scenarios
- Proactive vs. reactive AI: Examples in e-commerce, healthcare, and finance
- Mapping customer journeys to agentic AI capabilities
- Group Activity: Brainstorm a customer interaction problem agentic AI could solve
4. Setting the Stage
- Tools and platforms
- LangChain, Hugging Face, API frameworks, etc.
- Course project intro: Build an agentic AI customer assistant
- Hands-on Lab: Set up development environment (Python, API’s, LLM access)
Day Two: Technical Foundation and Building Blocks
5. LLMs and Reasoning
- How LLMs power natural language understanding and generation
- Prompt engineering for goal-oriented tasks
- Hands-on Lab: Create a basic LLM-powered Q&A system
6. Adding Context with RAG
- What is Retrieval-Augmented Generation?
- Connecting LLMs to external data sources
- Hands-On Lab: Build a RAG system to fetch real-time order status
7. Action Capabilities
- Integrating APIs for task execution
- Designing multi-step workflows
- Hands-on Lab: Add an API call to reschedule a mock delivery
8. Wrap-up and Project Work
- Combining Day 2 skills into a mini-agent
- Group Discussion: Challenges in autonomy and decision making
- Project Time: Begin Coding the customer assistant (i.e. order support agent
Day Three: Designing Autonomous Agents
9. Autonomy and Planning
- How agentic AI breaks down goals into actionable steps
- Algorithms for planning (i.e. tree search reinforcement learning basics
- Demo: An agent resolving a multi-step customer query
10. Memory and Context Management
- Short-term vs. long-term memory in AI Agents
- Maintaining conversation context across interactions
- Hands-on Lab: Add memory to your agent for follow-up questions
11. Scaling Agentic AI
- Handling high volumes of customer interactions
- Load balancing and performance optimization
- Case Study: A real-world deployment
12. Project Development
- Refine the customer assistant: Add planning and memory features
- Peer Review: Share progress and Troubleshoot
Day 4: Business Applications and Deployment
13. Business Use Cases
- Customer support: Ticketing, refunds, escalations
- Sales: Upselling, personalized recommendations
- Proactive care: Anticipating customer needs
- Group Activity: Design an agentic AI for a specific industry
14. Integration with Business Systems
- Connecting to CRMs (e.g., Salesforce), ERPs, and messaging platforms
- Security and data privacy considerations
- Hands-On: Simulate integration with a mock CRM
15. Deployment Strategies
- Cloud vs. on-premise deployment
- Monitoring and maintaining AI agents
- Demo: Deploy a sample agent to a cloud platform
16. Project Refinement
- Finalize the customer assistant: Test multi-step workflows
- Prepare for Day 5 presentations
Day 5: Ethics, Evaluation, and Capstone
17. Ethics and Governance
- Bias and fairness in autonomous decision-making
- Transparency: Letting customers know they’re interacting with AI
- Guardrails and human oversight
- Discussion: Ethical dilemmas in customer-facing AI
18. Evaluating Agentic AI
- Metrics: Accuracy, customer satisfaction, task completion rate
- Testing for edge cases and failure modes
- Hands-On: Test your agent with simulated customer scenarios
19. Capstone Project Presentations
- Teams present their customer assistants (e.g., demo + explanation)
- Feedback from peers and instructors
20. Wrap-Up and Next Steps
- Recap of key learnings
- Resources for further study (e.g., frameworks, research papers)
- Q&A and course feedback