Subscribe and Access : 5200+ FREE Videos and 21+ Subjects Like CRT, SoftSkills, JAVA, Hadoop, Microsoft .NET, Testing Tools etc..

Batch Date: Oct 6th @9:00PM

Faculty: Mr. Naveen Mourya (9+ Yrs of Exp,..)

Duration: 3.5 Months

Venue :
DURGA SOFTWARE SOLUTIONS,
Flat No : 202, 2nd Floor,
HUDA Maitrivanam,
Ameerpet, Hyderabad - 500038

Ph.No: +91 - 8885252627, 9246212143, 80 96 96 96 96

Syllabus:

GENERATIVE AI + AGENTIC AI

Module 1: Python, Maths & Statistics Fundamentals

Objective: Build strong Programming, Maths & Statistical fundamentals for AI and data processing.

1.1 Python Basics:

• What is Python and why is it popular in AI?
• Identifiers, Key Words, Basic Data types: (Int , Float, String, Complex & Boolean)
• Fundamental Data types: [List],(tuple),{set} & {dict}, Basic Syntax, Advantages.
• Operators, Operator precedence & Expressions

1.2 Control Flow:

• What are Control Flow Statements in Python? Advantages & Basic Syntax
• Conditional Statements ( if, if-elif, if-else & if-elif-else)
• Transfer Statements (break, continue & pass)
• Iterative Statements (for & while)

1.3 Functions & Modules:

• Functions: In built & User Defined Functions
• Parameters, Return Statements, Types of arguments & Variables
• Recursive/Nested/lambda & Syntax/ filter/map() with lambda & without lambda, reduce()
• Function Aliasing, import concept and Function vs Module vs Library
• Standard Modules: datetime, os, math, random, re, json, requests & use cases

1.4 File Handling & Exception Handling:

• What is File Handling: Importance in AI
• Types of Files: [Text Files, Binary Files], Opening & Closing a File, Reading Data from text files, Writing Data to text files.
• The with statement - (The seek() and tell() methods:)
• Handling csv files: [Writing data to csv file, Reading Data from csv file]
• Handling Json Files:[Writing data to Json file, Reading Data from Json file]
• Exception Handling: [What is Exception , Default Exception Handling in Python]
• REST APIs calling using Json, Import & Requests modules

1.5 Libraries:

NumPy (for tensors, embeddings, vector math)

• NumPy Basics: What is NumPy and why is it important in ML/AI, NumPy installation
• Arrays: 1D, 2D, and higher dimensions, Applications
• Difference between Python list and NumPy array
• Numpy Advanced: Array indexing and slicing, Mathematical operations, Reshape, flatten, Broadcasting
• Useful functions: arange(), linspace(), eye(), ones(), zeros()

Pandas (for cleaning and preparing tabular/nested data)

• Pandas Basics: What is Pandas and use cases
• Series and DataFrames, Creating DataFrames from dict/list/CSV, Viewing data: head(), tail(), info(), describe()
• Pandas Advanced: Indexing, slicing, filtering, Adding/deleting columns
• Aggregations: groupby(), sum(), mean()
• Handling missing values: isna(), fillna(), replace()

Matplotlib (for visualizing model performance & embeddings)

• Matplotlib Basics: Introduction to Matplotlib, pyplot and plotting syntax
• Line plots, bar plots, scatter plots with Titles, labels, legends
• Matplotlib Advanced: Subplots, Histograms, Pie charts with Styling: colors, markers, line types

1.6 Linear Algebra

• Vectors, Matrices, Matrix Operations
• Eigen Vectors & Eigenvalues

1.7 Probability & Statistics

• Probability, Conditional Probability & Distributions
• Statistical Measures (Z score, Skewness, Kurtosis, Geometric Distribution)
• Bias, Variance, Standard Deviation & Covariance
• Population, Sample, Data Types, Sampling Methods & Variables
• Measure of Central Tendency, Symmetry, Spread & Variability
• Hypothesis Testing (Null & Alternative Hypothesis, Type-I & II Errors)

Module 2: Introduction to Artificial Intelligence

Objective: Understand the foundations of AI, its history, and applications.

• What is Artificial Intelligence?
• History and Evolution of AI
• Types of AI: Narrow AI, General AI, and Superintelligent AI
• Applications of AI in Real World
• AI vs Machine Learning vs Deep Learning vs Generative AI

Module 3: Machine Learning Foundations

Objective: Build a strong foundation in ML concepts and techniques.

• Introduction to Machine Learning
• Types of Machine Learning: Supervised, Unsupervised, Reinforcement Learning
• Data Preprocessing: Cleaning, Normalization, Feature Engineering
• Train-Test Split, Cross-Validation
• Evaluation Metrics: Accuracy, Precision, Recall, F1-score, ROC-AUC
• ML Algorithms Overview:
  • Linear Regression, Logistic Regression
  • Decision Trees, Random Forests, Gradient Boosting
  • KNN, Naive Bayes, SVM
  • Clustering: K-means, Hierarchical
• Bias-Variance Tradeoff

Module 4: Deep Learning & Neural Networks

Objective: Build Deep Learning foundations, understand architectures, and apply them using TensorFlow and PyTorch.

4.1 Introduction & Foundations

• Introduction to Deep Learning
• History of Deep Learning
• Applications of Deep Learning

4.2 Neural Networks Basics

• Perceptron: Simple Neural Network Explained
• Activation Functions: Sigmoid, Tanh, ReLU, Leaky ReLU, Softmax
• Loss Functions:
  • Classification: Cross-Entropy Loss, Hinge Loss
  • Regression: Mean Squared Error (MSE), Mean Absolute Error (MAE), Huber Loss

4.3 Training Fundamentals

• Gradient Descent and Types: Batch, Stochastic, Mini-Batch
• Optimizers: SGD, Adam, RMSProp
• Parameters vs Hyperparameter

4.4 Frameworks & Practice

• Introduction to TensorFlow and PyTorch
• Hands-on exercises with both frameworks

4.5 Regularization & Model Generalization

• Concept of Underfitting and Overfitting
• Regularization Techniques: Dropout, Early Stopping, Batch Normalization, L1/L2 Regularization

4.6 Hardware for Deep Learning

• Introduction to GPUs
• Types of GPUs and their importance in Deep Learning

4.7 Projects

• Five practical Deep Learning projects covering classification, regression, image data, and text data

Module 5: Natural Language Processing (NLP)

Objective: Enable machines to understand and Process Human Language

5.1 Introduction to NLP

• What is NLP and why is it important?
• History of NLP and real-world applications

5.2 Text Pre-processing

• Lowercasing
• Punctuation Removal
• Stopword Removal
• Stemming
• Lemmatization
• POS Tagging (Parts of Speech Tagging)
• Named Entity Recognition (NER)

5.3 Text Vectorization

• Bag of Words (BoW)
• N-Grams
• TF-IDF (Term Frequency – Inverse Document Frequency)
• One-Hot Encoding

5.4 Word Embeddings

• Word2Vec
• GloVe
• FastText

5.5 Sequence-to-Sequence Models

• RNN (Recurrent Neural Networks)
• LSTM (Long Short-Term Memory)
• GRU (Gated Recurrent Unit)
• Encoder-Decoder Architecture explained
• Applications of Seq2Seq models (translation, summarization, chatbots, etc.)
• (Mini project and Exercises on this topic are by default covered)

Module 6: Transformers & Modern NLP

• Introduction to Transformers
• Attention Mechanism
• Self-Attention Explained
• Transformer Architecture (Encoder, Decoder)
• BERT, GPT, T5 and other Transformer Models
• Hugging Face Transformers Library
• Fine-tuning Pre-trained Models
• Applications of Transformers in NLP

Module 7: Generative AI & Large Language Models (LLMs)

Objective: Build strong foundations in Generative AI with a focus on text-based LLMs.

7.1 Introduction to Generative AI

• What is Generative AI?
• Why Generative AI matters in text-based applications
• Generative Models vs Discriminative Models

7.2 Large Language Models (LLMs)

• What is a Large Language Model?
• How LLMs are trained and work at a high level
• Categorization of LLMs (Open-source vs Proprietary, General-purpose vs Domain-specific, etc.)
• Popular LLM families: GPT, LLaMA, Falcon, Mistral, Claude, Gemini, etc.
• Companies providing LLMs (OpenAI, Meta, Anthropic, Google DeepMind, Mistral, Cohere, etc.)

7.3 Applications of Text-based Generative AI

• Chatbots & Conversational AI
• Text Summarization
• Text Generation (content creation, code generation, etc.)
• Q&A Systems

7.4 Practical Understanding

• Introduction to Tokenization and Embeddings (basic overview)

Module 8: Building Applications with LangChain & RAG

Objective: Learn how to build practical applications with LLMs using LangChain and Retrieval-Augmented Generation (RAG).

8.1 Creating Applications with LLMs

• How LLM-based applications are structured
• Connecting LLMs with external data and tools
• Designing interactive applications with LLMs

8.2 Introduction to LangChain

• What is LangChain and why do we need it?
• LangChain architecture and workflow

8.3 LangChain Components

• Chains: LLMChain, SimpleSequentialChain, SequentialChain, ConversationChain
• Prompt Templates: Standard, Few-shot, Zero-shot, and Custom templates
• Memory Types:
  • ConversationBufferMemory
  • ConversationSummaryMemory
  • ConversationBufferWindowMemory
  • VectorStoreRetrieverMemory
• Agents & Tools: How LangChain integrates tools for dynamic reasoning
• Document Loaders & Text Splitters: Preparing external knowledge for LLMs

8.4 Retrieval-Augmented Generation (RAG)

• What is RAG and why it’s needed
• Components of RAG:
  • Retriever
  • Vector Database (FAISS, Pinecone, Weaviate)
  • LLM Integration for answering queries
• Workflow of RAG (Step-by-step explanation)
• Applications of RAG in chatbots, search, and knowledge assistants

8.5 Hands-on Applications

• Build a simple Q&A chatbot with LangChain
• Implement RAG with a vector database

Module 9: LlamaIndex & Fine-Tuning of LLMs

Objective: Understand how to structure data pipelines with LlamaIndex and customize LLMs through fine-tuning techniques.

9.1 Introduction to LlamaIndex

• What is LlamaIndex?
• Why LlamaIndex is used in GenAI applications
• High-level architecture of LlamaIndex

9.2 Components of LlamaIndex

• Data Connectors (ingesting data from different sources)
• Indexes (Vector Index, List Index, Tree Index, Keyword Table Index)
• Query Engines (retrieval mechanisms)
• Storage Context & Persistence
• Integration with LLMs

9.3 Introduction to Fine-Tuning LLMs

• What is Fine-Tuning?
• Why fine-tuning is needed in enterprise and domain-specific contexts
• Fine-tuning vs Prompt Engineering

9.4 Types of Fine-Tuning

• Full Fine-Tuning: Updating all model parameters
• Partial Fine-Tuning (PEFT): Updating only small parameter-efficient layers
• Popular PEFT Techniques:
  • LoRA (Low-Rank Adaptation)
  • QLoRA (Quantized LoRA for memory-efficient fine-tuning

9.5 Practical Insights

• Trade-offs between full and partial fine-tuning
• Choosing the right fine-tuning strategy for your use case

Module 10: Capstone Projects on Cloud Platforms

Objective: Apply everything learned to real-world Generative AI projects deployed on cloud platforms.

10.1 Journey Recap

• By this stage, learners would have already completed 25+ hands-on projects across ML(2), DL(5), NLP(4), Transformers (4), LLMs-LangChain (8), RAG (6), Llama index (4) and Fine-tuning. (1) apart from the Exercises.
• Now we focus on end-to-end Capstone Projects hosted on cloud platforms.

10.2 Capstone Project on AWS (Amazon Web Services)

• Introduction to AWS Cloud
• Overview of AWS services relevant to AI/ML application building:
  • Compute: EC2, Lambda
  • Storage: S3
  • Other Integrations (API Gateway, CloudWatch)
• Capstone Project: Building and deploying a Generative AI Chatbot on AWS

10.3 Capstone Project on GCP (Google Cloud Platform)

• Introduction to Google Cloud
• Overview of GCP services relevant to AI/ML application building:
  • Compute: Compute Engine, Cloud Run
  • Storage: Cloud Storage (GCS)
  • Databases: BigQuery
  • AI/ML: Vertex AI
  • Other Integrations (Pub/Sub, Cloud Functions)
• Capstone Project: Building and deploying a Generative AI Chatbot on GCP

Module 11: Prompt Engineering

Objective: To enable users to effectively communicate with and control AI models to achieve desired outcomes.

• Core prompting techniques
  • Zero-shot prompting, Few-shot prompting, Chain-of-Thought (CoT) prompting, Tree-of-Thought (ToT) prompting
• Advanced prompting strategies
  • Persona-based prompting, Meta-prompting, Constraint-based prompting, Negative prompting, Retrieval-Augmented Generation (RAG)
• Practical applications and real-world tools
  • Text generation, Code assistance, Multimodal prompting, AI development platforms

Module 12: Agentic AI

Objective: Enabling systems to operate autonomously, make independent, complex decisions, adapt to changing environments, and solve multi-step problems with minimal human intervention.

• Introduction to Agentic AI
  • What is Agentic AI?
  • Core AI agent building blocks - Perception, Cognition/Reasoning, Planning, Action, Memory, Adaptability & learning:
    
• Agentic AI architectures and design patterns
  • Architectural concepts
  • Key design patterns - ReAct, Reflection, Tool Use & Human-in-the-Loop (HITL)
• Foundational frameworks and technologies
  • LangChain and LangGraph, Crew AI & Multi-agent systems:
• Practical agent development and deployment
  • Building agents with Python, Tools and APIs, No-code/low-code agent Development & Cloud deployment
    
• Responsible AI and evaluation
  • Observability and evaluation. Ethical considerations and risk mitigation
• Future of Agentic AI & AGI