Demystifying AI From Basics to Agents

Demystifying AI From Basics to Agents

Unlock the power of AI — from rule-based systems to autonomous agents — with applied insights and ethical clarity in a day

AI isn't coming—it's already here, quietly running the systems we use every day. While everyone else argues about the future, smart professionals are learning how these tools actually work under the hood.

This course strips away the buzzwords and marketing speak to reveal what's really happening in AI right now. You'll understand the mechanics behind machine learning, dive into the architecture of large language models, and explore how autonomous agents are reshaping entire industries. We'll tackle everything from Model Context Protocol to the ethical landmines that trip up even seasoned teams.

No fluff, no speculation. Just decades of real-world experience distilled into practical knowledge you can use immediately. Because understanding AI isn't about predicting the future—it's about navigating the present.

Learning Outcomes

By the end of the day, participants will be able to:

• Differentiate between types of AI (narrow, general, reactive, agentic)
• Explain the mechanics behind ML and DL, including model training and evaluation
• Understand how LLMs are built and used with prompting
• Describe the Model Context Protocol and its role in tool integration
• Grasp the concept, use cases, and risks of agentic AI
• Identify key ethical issues—bias, accountability, transparency—in AI systems

Prerequisites

• Basic familiarity with AI/ML terminology (e.g., “model,” “training,” “dataset”)
• Comfort with high-level concepts in programming, statistics, or data processing
• Openness to discuss real-world examples and ethical dilemmas

Detailed Course Topics

1. Types of AI (Foundations)
   1. Reactive vs. Deliberative vs. Learning agents
   2. Narrow AI vs. Artificial General Intelligence (AGI)
   3. Agentic AI as an emerging class
2. Machine Learning & Deep Learning
   1. Overview: supervised, unsupervised, reinforcement learning
   2. Deep neural networks, architectures (CNNs, RNNs), backpropagation
   3. Use of Deep Learning in agentic systems
3. Large Language Models (LLMs) & Prompting
   1. What makes LLMs 'large': data, transformer architecture, token prediction
   2. Prompt design: re-prompting, context windows, chain‑of‑thought
   3. Integrating structured prompting with MCP
4. Model Context Protocol (MCP)
   1. Origin and purpose: Anthropic’s 2024 open‑source standard
   2. MCP capabilities: file access, function calls, multi-tool orchestration
   3. Adoption by OpenAI, DeepMind, Microsoft and benefits/interoperability
   4. Security considerations: data permissions, prompt injection vulnerabilities
5. Agentic AI (Autonomous Agents)
   1. Definition: systems that perceive, reason, plan, act with minimal oversight
   2. Core traits: autonomy, reasoning, planning, context awareness
   3. Business & tech trends: mainstream enterprise pilot rates, Microsoft/OpenAI momentum
   4. Real-world use cases:
      1. Digital workers (BNY, Intuit QuickBooks, Salesforce Agentforce)
      2. Cybersecurity, finance, supply chain, customer service automation
   5. Risks & caveats:
      1. Business model uncertainty
      2. Autonomy plus accountability: incidents in blackmail & unauthorized actions
6. AI Ethics (Governance & Responsibility)
   1. Key principles: fairness, transparency, privacy, accountability
   2. Machine ethics taxonomy: impact, implicit, explicit, full agents
   3. Explainable AI (XAI): transparency in black‑box systems
   4. Governance issues in agentic AI: oversight, human‑in‑the‑loop, audit tracing
   5. Broader risks: job displacement, misinformation, alignment with human values
7. Synthesis & Discussion
   1. How ML/DL, LLMs, MCP, and ethics combine in end‑to‑end agent systems
   2. Group activity: draft a responsible agent proposal (use case, prompt, tools, safe‑guards)
   3. Practical Q\&A: deployment, monitoring, future-proofing

This structure blends technical depth with real-world examples, ensuring participants leave with a grounded yet comprehensive understanding of modern AI—from fundamental paradigms to autonomous systems and ethical guardrails.