The Fifth Elephant
The Fifth Elephant

AI, analytics and data workflows - bi-monthly meet-ups

Call for submissions

All submissions

Powered by VideoKen

Swetha A

Swetha A

@swetha_03

Mahita D

Mahita D

@mahita07

Crafting Multi-Agent Systems That Think and Act

Submitted Dec 30, 2025

Type of submission: Hands-on workshop (1.5 hrs to 3 hrs)

Abstract

In this hands-on workshop, you’ll learn how to design and orchestrate intelligent AI systems using LangGraph and modern agentic patterns. You’ll start by understanding the building blocks such as nodes, edges, workflows, and message routing. From there, you’ll dive into agent architecture: how prompts, tools and context come together to enable reasoning-driven automation.
You’ll also explore multi-agentic patterns, learning when and how to coordinate multiple agents to solve complex, real-world problems. Finally, you’ll work with the Model Context Protocol (MCP) to externalize knowledge and connect agents to structured data and services.
Through guided exercises and practical implementation patterns, you’ll leave with the skills to build multi-agentic systems, transforming LLMs from chat interfaces into powerful, composable systems that integrate seamlessly across your engineering stack.

Workshop Agenda

1) Foundations of LangGraph — 30 mins

Objective: Understand the core building blocks of LangGraph
Topics:

  • What is a Graph?
  • Nodes, Edges & Execution Flow
  • Designing a Basic Workflow
  • Conditional Edges & Routing Logic
  • Message Types (System, User, AI)

Hands-On: Building Your First Workflow — 30 mins

Activity:
Create a simple multi-node LangGraph workflow from scratch

2) Introduction to AI Agents — 30 mins

Objective: Learn what agents are and how they operate within workflows
Topics:

  • What are AI Agents?
  • Prompt Engineering
  • Context Management
  • Tool Use & Execution Flow

Hands-On: Build a Simple Agent with Tools — 30 mins

Activity:
Design and run an agent that uses tools to solve a task

3) Multi-Agent Systems — 45 mins

Objective: Explore collaboration between multiple agents
Topics:

  • What is a Multi-Agent System?
  • Common Multi-Agent Patterns & Architectures
  • Designing Multi-Agentic Systems

Hands-On: Build a Multi-Agent System — 60 mins

Activity:
Create a multi-agent solution for a real-world use case

4) Model Context Protocol (MCP) — 30 mins

Objective: Understand how MCP extends agent capabilities
Topics:

  • What is MCP?
  • When & Why to Use MCP?
  • Integrating MCP with Agentic Systems

Hands-On: MCP Client Integration — 30 mins

Activity:
Build an MCP client and connect it to the agents

5) Q&A + Open Discussion — 15 mins

The workshop expands on the key concepts presented in this talk:
https://youtu.be/zWWWi5tfkn4?si=SXbpzCFN62GDcAxa

Pre-Requisites

  • Bring your own laptop
  • Working knowledge of Python is required
  • Git installed and an active GitHub account for code access and collaboration
  • Valid API keys provisioned in advance (e.g., Gemini, GPT).
    Note: API keys will not be provided during the workshop.
  • Familiarity with an IDE or code editor (VS Code, PyCharm, etc.)

Key Takeaways

  • Understand the LangGraph execution model, including node orchestration, edge routing, message propagation, and workflow lifecycle management.
  • Gain a deep understanding of agent architecture, including prompt composition, context scoping and tool invocation patterns.
  • Develop multi-agent systems, applying coordination patterns such as task delegation, routing, and role-based specialization.
  • Integrate Model Context Protocol (MCP) to externalize data access and extend agent capabilities through standardized context interfaces.

Target Audience

  • Developers and engineers building AI-powered applications who want to design reliable, agent-driven workflows
  • ML/AI engineers exploring agentic architectures
  • Tech leads and solution architects designing scalable, multi-agent systems and automation platforms
  • Platform and backend engineers integrating LLMs with real-world systems, APIs, and infrastructure
  • Innovation, R&D, and product teams experimenting with AI agents, orchestration frameworks, and MCP-based extensibility

Worshop Facilitators Bio

Swetha A

Swetha is a software developer and GenAI enthusiast working as a Solution Consultant at Sahaj Software. She is passionate about building intelligent systems that harness deep learning and generative AI to solve real-world challenges. With hands-on experience in AI-driven projects and a research paper presented at the International Conference on Data Analytics and Management, she brings curiosity, innovation, and technical depth to every problem she tackles. She has delivered talks on AI agents and facilitated workshops on AI-assisted SDLC, empowering people to apply AI thoughtfully.

LinkedIn: https://www.linkedin.com/in/swetha0302

Workshop conducted

Hasgeek Workshop conducted in MathCo Bangalore on Mastering Prompt Engineering across SDLC http://has.gy/v1bT

Talks on AI Agents

1)https://sahaj.ai/events/under-the-hood-of-agentic-ai/
2)https://youtu.be/zWWWi5tfkn4?si=SXbpzCFN62GDcAxa

Mahita D

Mahita is a full-stack developer passionate in agentic systems and the evolving role of AI in software development. She enjoys breaking down complex ideas into clear, approachable ideas. Her work focuses on exploring innovative ways AI can simplify complex problems and address everyday challenges. Through her talks, she makes working with AI systems more approachable and practical.

LinkedIn: https://www.linkedin.com/in/mahita07

https://sahaj.ai/events/ai-search-and-kubernetes-operators-explained/

Talks on AI Agents

1)https://sahaj.ai/events/under-the-hood-of-agentic-ai/
2)https://youtu.be/zWWWi5tfkn4?si=SXbpzCFN62GDcAxa

All submissions

Comments

{{ gettext('Login to leave a comment') }}

{{ gettext('Post a comment…') }}
{{ gettext('New comment') }}
{{ formTitle }}

{{ errorMsg }}

{{ gettext('No comments posted yet') }}

Hosted by

The Fifth Elephant
The Fifth Elephant
Jumpstart better data engineering and AI futures