Course Details
Course Code: – ;Â Â Duration: 3 Days; Instructor-led
This intensive program is designed to demystify the mechanics of Large Language Models (LLMs) and empower both technical and non-technical personnel. Participants will progress from understanding core AI principles to building and deploying multi-step, autonomous workflows using accessible no-code tools.
This course is not associated with any certification
Audience
Anyone who would like to use GenAI to build a workflow automation
Prerequisites
None
Methodology
We utilize a High-Engagement Model to ensure skills are retained:
- Interactive Facilitation: Briefings followed by immediate guided application.
- Sandbox Building: Hands-on time with no-code tools (n8n, Stack-AI, Zapier).
- Case-Based Logic: Solving real business friction points through agentic design.
Course Objectives
- Stage 1 (Foundations): Demystifying LLMs, mastering advanced prompting, and understanding Retrieval Augmented Generation (RAG).
- Stage 2 (Practitioner Workshop): Prototyping, designing, and strategically deploying autonomous workflows while navigating ethical and ROI considerations.
Outlines
- Quick History : Evolution of LLMs (GPT-3 to today’s multi-modal/ agentic LLMs).
- LLM Essentials: Types of LLMs, functions, and the concept of Token Size (the agent’s working memory).
- Prompt Engineering Fundamentals: Clarity, context, and constraints. The diAerence between a simple query and a goal-driven prompt.
- System Prompt Modification: Giving the agent a “Job Description” to define its personality and rules (non-technical intro to fine-tuning concept).
- LLM Limitation Mitigation: Introduction to embedding and vector databases.
- Data Quality: Simplified concepts of chunking and overlap size
- Basic RAG Concept: How agents look up private/current data before answering.
- Hands-On RAG Workflow: Demonstration and use case walkthrough of a simple RAG application using no-code platforms (e.g., n8n, stack-ai. com) to answer questions based on a specific document.
- Building Agents to Solve Problems: Focused examples on creating agents to improve business operations (e.g., automated report summarizing) and personal daily lives (e.g., organizing health information or managing a budget).
- Autonomous Workflow Design: Sense →Think →Act →Learn
- Framework : Deciding when to use AI vs. traditional automation (ROI/Data sensitivity)
- Low Code / No Code : Zapier, Make, and specialized AI orchestration tools.
- Self-Hosting: Exploring open-source environments for local, secure testing
- Hands-On Lab : Building a complex, multi-step agent flow.
- Integrating internal/external data: Integrate agent flow with internal / external data.
- Identifying reputational risks and hallucination safeguards
- Quantifying “Time Saved” vs. “Implementation Cost.
- Addressing bias, fairness, job impact, and transparency
- Governance, Monitoring, and Scaling:
- Reviewing agent logs
- setting performance metrics, and establishing security best practices for scaled deployment
- Learning Method: Hands-on, with guided facilitation
