Day 1 Module Details

Module 1 — The Age of Agentic AI: A high-energy keynote-style opening establishing the ‘why now’ urgency of Agentic AI. Covers the evolution of automation — from rule-based scripts to chatbots to copilots to fully autonomous agents — with real-world data and a dramatic live comparison demo.

• Live Audience Poll: ‘Is your organization AI-reactive or AI-proactive?’ using Mentimeter
• Video reel: 90 seconds of real AI agents in action (customer service, logistics, healthcare)
• Myth Busters: Facilitator debunks 5 common Agentic AI misconceptions with evidence

Module 2 — Agent Architecture Deep Dive : A structured technical walkthrough of how agents are built — explained through rich visuals and the analogy of a ‘virtual employee’ who perceives, thinks, plans, acts, and learns. Facilitator builds a live architecture diagram on the whiteboard.

• Framework Bingo: Participants match AI frameworks (LangChain, AutoGen, CrewAI) to use cases in a rapid-fire game
• Architecture sketch: Facilitator draws an end-to-end agent diagram live

Module 3 — Customer Journey AI Mapping: A structured group workshop using a pre-designed Customer Journey AI Canvas. Teams select an industry (banking, retail, healthcare, or telecom) and map customer pain points to potential agentic AI solutions. This canvas becomes the foundation for the Capstone Project.

• Gallery Walk: Teams display canvases; peer voting on most creative and most impactful AI use case
• Deliverable: Each team’s completed Customer Journey AI Canvas

Module 4 — Hands-On Lab 1: First Customer Agent: The first deep hands-on lab. Participants follow a guided workbook to build, test, and refine a working customer service agent. Platform: Flowise (drag-and-drop LangChain interface) — browser-based, no installation required.

• Lab Goal: A functional agent that handles FAQs, retrieves order data from a mock API, and escalates to a human when confidence is low
• Challenge Mode: Bonus tasks for faster participants — add a second language or a sarcasm-detection

Modules 5 & 6 — Prompting, Personas, Tools & Memory: Back-to-back practical labs focusing on the human feel and intelligence of agents. Participants craft persona-driven system prompts, then supercharge their agents with tool integrations and persistent memory.

• Prompt Battle: Two teams craft competing prompts; audience rates which agent performs better on 3 live test queries
• Memory Lab: Agent ‘remembers’ a returning customer across 5 conversation turns — participants experience the difference firsthand

Module 7 — Industry Case Studies: Three rapid-fire case studies presented in a panel/storytelling format. Each case covers the business problem, the agentic solution deployed, measurable outcomes achieved, and key lessons learned.

1. Case 1 — Islamic Banking: Digital onboarding agent achieving 78% reduction in processing time
2. Case 2 — Retail: Hyper-personalization agent delivering 34% uplift in average order value
3. Case 3 — Telecom: Proactive churn prevention agent reducing churn by 22% within 6 months

Day 2 Module Details

Module 8 — Multi-Agent Orchestration: The most technically ambitious session of the program. A lecture-demo hybrid where participants see how multiple specialized agents cooperate, hand off tasks, and resolve conflicts — illustrated through a live 3-agent customer complaint resolution demonstration.

1. Design Cards: Participants receive ‘Agent Role’ cards (Classifier, Retriever, Resolver, Escalator, Communicator) to use in Labs
2. Anti-Patterns: What goes wrong with poorly orchestrated agents — cascading failures, infinite loops, conflicting outputs
3. Live Demo: Full 3-agent pipeline resolves a banking dispute in real time on the screen

Module 9 — Hands-On Lab 2: Agent Teams: The most complex lab of the program. Participants build a 3-agent pipeline from scratch using CrewAI or equivalent. The pipeline handles customer complaints end-to-end — from intake classification through resolution to automated follow-up.

1. 5 Test Scenarios: angry customer, duplicate complaint, out-of-policy request, VIP customer, unclear/ambiguous query
2. Debrief: What broke? What surprised you? How would you redesign it?

Module 10 — Enterprise Integration Lab: A guided lab connecting agents to simulated enterprise APIs. Participants experience firsthand the challenge and reward of wiring AI agents to live data sources and communication channels.

1. Simulated CRM: Agent looks up customer tier, purchase history, and open support tickets
2. Ticketing System: Agent creates a ticket with correct category, priority, and team assignment
3. Communication Channel: Agent sends a templated WhatsApp/SMS message (mocked) with real order update data

Module 11 — Responsible AI & Guardrails: An interactive workshop using the Responsible AI Governance Canvas. Small groups identify where their customer-facing agent could fail — technically, ethically, regulatory, or operationally — then design mitigations.

1. Red Team Roleplay: One participant ‘attacks’ the agent with adversarial prompts; the group observes and designs guardrails
2. Governance Checklist Handout: 12-point pre-deployment responsible AI checklist
3. Regulatory overview: EU AI Act, Pakistan PDPA, and sector-specific AI guidelines

Module 12 — Monitoring & Observability: Shifts focus from building to running AI agents in production. Participants review a sample dashboard, identify red flags, and discuss incident response protocols when an agent is underperforming or producing harmful outputs.

• Dashboard Lab: Interpret 5 monitoring indicators and diagnose a specific underperforming agent
• Incident Response Simulation: Agent is hallucinating on a product return query — what do you do next?

Module 13 — ROI & Business Case: A focused, numbers-driven session helping participants translate AI agent value into language that resonates with finance and leadership teams. Uses a provided pre-built ROI Calculator template.

• ROI Calculator Exercise: Plug in your industry baseline metrics and see estimated impact in cost, CSAT, and revenue
• Real benchmarks: cost-per-contact saved, NPS delta, resolution time improvement, churn reduction rates

Module 14 — AI Adoption Roadmap: The strategic capstone before presentations. Participants complete a 1-page roadmap template they can present to stakeholders upon return to work

• RACE Framework: Readiness Assessment → Agent Selection → Connect to Systems → Expand & Scale
• Peer Review: Partners exchange roadmaps and provide 2 minutes of structured, rubric-guided feedback

Capstone Project Presentations: Teams of 3–4 participants present their complete Agentic AI solution, designed iteratively across both days. Each team receives 8 minutes to present plus 2 minutes for Q&A.

• Presentation covers: Problem Statement, Agent Architecture, Tools & Integration, Expected Outcomes, Governance Plan, ROI Projection
• Rubric: Problem clarity (20%) |  Agent design quality (25%)  |  Business impact (20%)  |  Governance awareness (15%)  |  Presentation clarity (20%)
• Awards: Best Overall Solution |  Most Creative Use Case  |  Strongest Business Case