Advanced Engineering Program in AI Agent Workflows and Agentic Systems Development

Program Description

The programme is designed to equip professionals with the skills to design, implement, and manage Agentic AI systems and workflows. It focuses on building intelligent, autonomous systems that can plan, execute, and optimize tasks with minimal human supervision, bridging the gap between AI theory and practical applications.

Key Highlights

check svg

On successful completion of the programme, you will receive a Certificate from IITM Pravartak and acquire competencies in product strategy, user-centered design, and innovation management, enabling advancement into senior product roles.

check svg

Optional 3-Day Campus Immersion at IITM Pravartak for collaborative, hands-on learning

Learning Format

Online

Duration

28 Weeks
Weekend live sessions (Saturday and Sunday)

Certified by

IITM Pravartak Technologies Foundation
Technology Innovation Hub (TIH) of IIT Madras and
Futurense

Program Fee

INR 1,25,000 + 18% GST

Education Qualification

Bachelor’s degree in Engineering, Computer Science, Mathematics, or related disciplines. Professionals with relevant industry experience in AI, ML, or automation are encouraged to apply.

Suggested Prerequisites

Foundational understanding of Artificial Intelligence, Machine Learning, and programming (preferably Python).

Teaching Hours

140 + Hours

Lead Faculty

IIT Madras Faculty and Industry Experts

Learning Module

  • Overview of Deep Learning and Neural Networks
  • Structure of Neural Networks and Backpropagation
  • Key Activation Functions (ReLU, Sigmoid, Tanh)
  • Training Neural Networks and Hyperparameter Tuning
  • Introduction to NLP and its Connection with Al Agents

  • NLP Techniques: Tokenization, Stemming, Lemmatization
  • TF-IDF
  • Word Embeddings: Word2Vec, GloVe
  • Generative Al: GANs, VAEs
  • Applications of Generative Models in Content Creation (text, art, music) Introduction to Transformers and GPT

  • Introduction to Large Language Models (LLMs)
  • Transformer Architecture and Attention Mechanisms
  • Pre-training and Fine-tuning LLMs
  • Use Cases: GPT, BERT, T5, and more
  • LLMs vs. Traditional Rule-based Systems

  • Understanding Al Agents and their Evolution Key
  • Differences between Al Models and Agents
  • Using Neural Networks and LLMs in Agent Design
  • Contextual Decision-making for Agents
  • Al Agent Frameworks and Tools

  • Holistic Analysis of Interconnected Workflows Identifying
  • Feedback Loops
  • Mapping Dependencies Emergent Behavior in MAS
  • Balancing Optimization Across Subsystems

 

  • Problem Immersion
  • Journey Mapping (As-ls➔ To-Be)
  • Leverage Point Discovery
  • KPI & Guardrail Definition
  • Service Blueprint Creation

  • Hierarchical
  • Manager-worker
  • Peer-to-peer
  • Blackboard
  • Contract-net
  • Debate/hybrid Human-agent

  • Understanding Context in Al Workflows
  • Passing Context between Agents
  • Context Engineering for Task Completion
  • Designing Inputs/outputs for Multi-agent Workflows

  • Multi-agent System Architecture
  • Agent Coordination and Task Delegation
  • Message Passing between Agents
  • Defining Agent Roles within Workflows

  • Workflow Design Templates
  • Task Decomposition and Agent Interaction
  • Process Mapping and Optimization
  • Designing Scalable Workflows for Complex Systems

  • Design-first Approach to Multi-agent Systems
  • Integrating Agent Workflow Design into Prototyping
  • Connecting Agents within Workflows to Simulate Real-world Applications

  • Integrating Real-world Tools with Agentic Workflows
  • Leveraging Memory Systems
  • Using External APls and Data Sources in Multi-agent Workflows
  • Scaling Workflows with Advanced Tools

  • Dynamic Input Handling
  • Context Injection Techniques for Adaptive Agent Workflows
  • Role-based Prompts and Task-specific Agent Communication

  • Business-driven Multi-agent System Automation
  • Automating Complex Business Workflows with Multiple Agents
  • Ensuring Smooth Coordination and Context Passing

  • Understanding Real-time Monitoring of Multi-agent Workflows
  • Using Logging and Monitoring Tools to Track Agent Performance
  • Debugging Agent Communication and Task Delegation Errors

  • Integrating Agent Systems into Business Enterprise Workflows
  • Bridging Agents with Enterprise Resource Planning (ERP) and Customer Relationship Management (CRM) Systems

  • Deep Dive into Context Engineering for Multi-agent Systems
  • Handling Complex Context Switches in Large-scale Workflows
  • Adaptive Task Handling


Are you interested in this program?

Our Learning Partners

Want To Know More

Guiding Star with Our Help!

Contact Us