Understanding AI and Tools

Understanding AI and Tools

From Concepts to Tools for Analysts in a day

AI is no longer the domain of programmers and researchers alone. For analysts and business professionals, an informed grasp of artificial intelligence — what it is, how it works, and which tools can amplify your impact — is rapidly becoming a critical skill. In this one-day workshop, an instructor with over 30 years of industry experience — not an academic theorist — presents real-world, demand-driven content. You’ll explore AI foundations, types of systems, and practical tools you can use today to enhance your analytical capabilities and business outcomes.

Learning Outcomes

By the end of this course, participants will be able to:

• Explain what artificial intelligence (AI) is, how it works (including core mechanisms), and distinguish its main types (e.g., rule‐based, machine learning, large language models)
• Describe what large-language models (LLMs), agentic AI systems and the emerging protocol frameworks (e.g., the Model Context Protocol – MCP) are, and why they matter for analytics and business use
• Navigate and compare key AI tools suitable for non-developers, including enterprise platforms (e.g., Microsoft Copilot / AI Hub for the Power Platform), conversational assistants (e.g., ChatGPT), agentic tools (e.g., Manus AI) and learning/creation platforms (e.g., NotebookLM)
• Assess how to apply these tools in their analyst roles – identifying opportunities, selecting appropriate tools, understanding limitations and knowing the risks
• Form a practical action plan for experimentation and integration of AI capabilities into their analytical workflows

Prerequisites

Participants should have:

• A solid comfort level with analytics or business-intelligence work (e.g., data interpretation, reporting, dashboarding)
• Familiarity with standard business software (spreadsheets, databases, dashboard tools)
• Basic understanding of what “cloud tools” and “software as a service (SaaS)” mean – though no programming or data-science coding experience is required
• A curious mindset and willingness to explore new tools and think about process change

Training Outline

Below is a detailed outline of the course content, organised by major topic areas and sub-topics.

1. Foundations of Artificial Intelligence

• Defining AI: What it is, what it isn’t
• How AI works: overview of core mechanisms
  • Rule-based systems / expert systems
  • Machine learning (supervised, unsupervised, reinforcement)
  • Deep learning and neural networks
• Introduction to Large Language Models (LLMs)
  • What an LLM is (architectural overview)
  • Training data, fine-tuning, inference
  • Strengths and limitations (e.g., language generation, reasoning boundaries)
• Types / categories of AI relevant to analysts
  • Generative AI (text, image, audio)
  • Predictive analytics and prescriptive AI
  • Conversational‐AI and assistants
  • Agentic AI: autonomous systems that act and decide
• The impact of AI on analytics / decision-making workflows
  • Automation of routine tasks
  • Augmentation of insight discovery
  • Risk of over-reliance, bias and “hallucination”

2. Emerging Frameworks and Architectures:
Agentic AI & MCP

• What is an AI “agent”?
  • Definition: systems that observe, plan, act, adapt.
  • Difference between “just a chatbot/LLM” vs agentic system
• The role of protocols in AI integration: Model Context Protocol (MCP)
  • What MCP is: open standard to enable AI agents to plug into tools and data sources.
  • Why it matters: enables interoperability, reduces custom integrations.
  • Implications for business / analytics: bridging the gap between models, data and workflows
• Agentic frameworks in practice: how analytics teams might benefit
  • Use-cases: multi-step workflows, tool orchestration, self-service automation
  • Limitations, governance, risk considerations (data, security, auditability)

3. Key AI Tool Categories for Analysts & Non-Developers

• Conversational assistants / generative AI platforms
  • Chatbots such as ChatGPT: what they can do, features for analysts (e.g., summarisation, generation, Q&A)
  • Special features: e.g., plugins, “function calling”, enterprise versions
• Enterprise AI hubs / platforms
  • Microsoft’s AI Hub for the Power Platform: illustrating the incorporation of AI capabilities in business apps and low-code environments
  • How these platforms enable non-developers to access and apply AI (via GUIs, connectors, dashboards)
• Agentic tools / workflow automation platforms
  • Tools that go beyond chat to orchestrate tasks, call data sources, drive actions (e.g., Manus AI)
  • How these might change how analysts perform tasks, collaborate with AI “agents”
• Learning & creation platforms
  • Platforms that empower knowledge work, learning, research support (e.g., NotebookLM)
  • Use-cases in analyst roles: document summarisation, insight amplification, knowledge management

4. Tool Explorations – Hands-On Review & Comparison

• Platform A: Microsoft AI Hub / Copilot for Power Platform
  • Overview of capabilities for analysts: data ingestion, workflow automation, chatbot integration, dashboards
  • Discussion of business-use cases in analytics teams
  • Considerations: adoption, governance, change management
• Platform B: ChatGPT (special features relevant for analysts)
  • Review of features: prompt engineering for analytics tasks, plugins/extensions, data summarisation, code generation assistance (for non-dev use)
  • Practical tips: framing prompts, verification of outputs, integrating into existing workflows
• Platform C: Agentic offering – Manus AI
  • What Manus AI is: a multi-agent system that executes tasks, automates workflows.
  • Discussion: when/why such agentic tools matter for analytics teams
  • Challenges: readiness, reliability, risk, supervision
• Platform D: Learning & Creation – NotebookLM
  • What NotebookLM is: an AI research/knowledge assistant from Google, accepts uploads of PDFs, websites, videos, outputs summarisation, Q&A.
  • Use-cases for analysts: synthesising research, building briefing documents, knowledge repositories
  • Best practices: source verification, limitation awareness

5. Application, Opportunity & Action Planning

• Identifying opportunity spaces in analyst workflows
  • Routine tasks that can be automated or enhanced (e.g., data cleaning, report summarisation, insight generation)
  • Value-add tasks where AI augments human judgment (e.g., scenario modelling, narrative building)
• Tool-selection criteria: matching task to tool
  • Checklist: ease of use, data connectivity, governance/security, cost, scalability
• Risk & governance considerations
  • Data privacy, accuracy, bias, audit trail, human oversight
  • Change management: training, user adoption, role evolution
• Constructing a pilot roadmap
  • Define objective, scope, success criteria
  • Select tool(s), assign roles, measure outcomes
  • Planning for scale and sustainability
• Wrap-up: Q&A and next steps for participants
  • Encouraging participants to identify at least one experiment to try within 30 days
  • Resources for further learning and community inclusion

This single-day course is designed to move beyond theoretical AI and into actionable insight for analysts and business professionals. The instructor’s 30+ years of industry experience ensures practical, applicable content — no academic fluff.