AI & Agentic AI Training Program

Curriculum — 6 Phases

A comprehensive journey from foundational ML to advanced agentic AI systems.

Phase 1

Core AI & ML Foundations

Mathematics for AI

Linear Algebra, Probability, Statistics, Optimization, Gradient Descent

NumPy SciPy

Python for Machine Learning

Data manipulation, visualization, feature engineering

Pandas Matplotlib Scikit-Learn

Core ML Algorithms

Regression, Classification, Clustering, Decision Trees, Random Forest, XGBoost

Scikit-Learn

Deep Learning Essentials

Neural networks, CNN, RNN, LSTM, Transformers (intro)

TensorFlow PyTorch

Model Lifecycle Management

Training, validation, testing, model evaluation metrics, hyperparameter tuning

MLflow DVC

Phase 2

Generative AI & Large Language Models

Foundation Model Concepts

What are LLMs, Tokenization, Transformer Architecture, Self-Attention

GPT LLaMA Claude Gemini

Prompt Engineering

Instruction prompting, few-shot learning, role-based prompting, chain-of-thought

OpenAI Playground PromptLayer

Customizing LLMs

Fine-tuning, LoRA, prompt-tuning, embeddings, adapter layers

HuggingFace OpenAI

LangChain Basics

Introduction, Chains, Agents, Tools, Memory, Callbacks, Streaming responses

LangChain

Vector Databases

Embedding creation, similarity search, storage, and retrieval

FAISS Pinecone Chroma

Phase 3

Retrieval-Augmented Generation (RAG) Systems

RAG Architecture Overview

Why RAG is needed, how it improves accuracy and grounding

Conceptual frameworks

Knowledge Store Design

Chunking, embedding, retrieval, re-ranking, indexing

FAISS Pinecone Chroma

Integrating RAG with LLMs

Context injection, query optimization, hallucination control

LangChain OpenAI API

Advanced RAG Use Cases

Enterprise knowledge base search, document Q&A, conversational RAG

Custom APIs

Enterprise Implementation

Secured RAG pipelines, compliance, latency optimization

AWS Bedrock GCP Vertex AI

Phase 4

Agentic AI — Multi-Agent Systems & Orchestration

Introduction to Agentic AI

Evolution from LLM apps → multi-agent ecosystems

Conceptual overview

AI Agent Architecture

Planner, Executor, Memory, Tool Interface, Communication Layer

CrewAI AutoGen LangGraph

LangGraph Deep Dive

Graph-based agent orchestration, task flows, message passing, state management

LangGraph

CrewAI / AutoGen Overview

Multi-agent collaboration, task delegation, error handling

CrewAI AutoGen

Model Context Protocol (MCP)

Tool registration, context passing, execution orchestration

MCP Frameworks

Orchestrator Implementation

Multi-agent coordination across LLMs, APIs, voice & data layers

Custom SDKs LangGraph + MCP

Real-World Integration

Genesys, AWS Connect, NovaSonic, Twilio, and enterprise connectors

Intellectt Universal Connector

Phase 5

MLOps, Deployment, and Scaling

MLOps Fundamentals

CI/CD for ML, model versioning, monitoring, A/B testing

Docker MLflow Kubeflow

Cloud Deployment

Containerization, Fargate, EKS, Lambda orchestration

AWS Azure GCP

LLM & Agent Deployment

Serving LLMs securely, latency control, API gateways

FastAPI LangServe

Enterprise-Grade DR & Failover

Designing fault-tolerant AI systems

AWS Multi-AZ

Phase 6

AI Governance, Ethics & Compliance

Responsible AI Practices

Model interpretability and explainability

SHAP LIME Fairlearn

Data Privacy & Compliance

Data privacy, IP ownership, and audit frameworks

GDPR Tools CCPA Tools

Security & Access Management

Secure API key and access management for multi-agent setups

Vault AWS Secrets Manager

Hands-on Projects & Capstone

Apply your skills through real-world projects and a comprehensive capstone.

Conversational RAG Bot

Build a context-aware chatbot using RAG architecture with vector databases.

Research Assistant Agent

Create an autonomous agent that researches, summarizes, and synthesizes information.

AI Workflow Automator

Design multi-agent systems that automate complex business workflows.

Speech-to-Speech Orchestrator

Build end-to-end voice AI systems with speech recognition and synthesis.

Capstone Project

Enterprise Voice Orchestrator

Design and deploy a production-ready multi-agent voice AI system that handles complex customer interactions, integrates with enterprise tools, and scales to handle real-world traffic. This capstone brings together everything you've learned across all 6 phases.

Duration 2 Weeks

Team Size 2-3 Members

Deliverable Production Demo

Eligibility & Assessment

Join a cohort of motivated learners ready to master AI and agentic systems.

Eligibility

Strong Python programming skills
Basic understanding of machine learning concepts
Familiarity with data structures and algorithms
Bachelor's degree in CS, Engineering, or related field or equivalent experience
Passion for AI and willingness to learn

Candidates with non-traditional backgrounds are encouraged to apply. We value diverse perspectives and experiences.

Assessment Process

Application Review

Submit your application with resume and statement of purpose

Technical Assessment

Complete a Python and ML fundamentals assessment (60 mins)

Interview

30-minute technical interview with program instructors

Admission Decision

Receive decision within 5 business days