AI and Data Science Course in Pune

Machine Learning Algorithms: 

• Supervised Machine Learning: 
• Linear Regression 
• Logistic Regression 
• K-Nearest Neighbour 
• Decision Tree 
• Random Forest 
• AdaBoost 
• Naive Bayes Classifier 
• Gradient Boost 
• Unsupervised Machine Learning:
• K-Means Clustering 
• Principal Component Analysis

Data Preprocessing:

• Handling of Missing Values 
• Handling of Outliers: 
• Detection(Zscore, IQR) 
• Handling (Imputation, Transformation) 
• Encoding(Label Encoding, One Hot Encoding) 
• Feature scaling(Normalisation,Standardization) 
• Binning

Feature Selection: 

• Filter Method 
• Wrapper Method 
• Embedded Method 

Hypothesis Testing: 

• P-value 
• Z-Test, T-Test 
• ANOVA Test, Chi-Square Test
• Shapiro

Introduction to DL

• DL vs ML
• Types of neural networks ? 
• DL applications
• Perceptron

Structure and learning process in neural networks

• Neural network
• Structure and data flow in neural network
• Training of neural networks

Types of function used in neural network

• Optimization function
• Batch Gradient Descent
• Stochastic Gradient Descent
• Mini Batch Stochastic Gradient Descent
• Mini Batch Stochastic Gradient Descent with Momentum
• Ada Grad
• Ada Delta
• Adam
• Activation functions
• Sigmoid, Tanh, Softmax, Relu, Leaky ReLu and Elu
• Loss functions
• MSE, MAE, MSLE
• Binary cross entropy
• Hinge loss
• Categorical cross entropy loss function
• Sparse categorical cross entropy loss function

Implementation of ANN

• Scripting of neural networks
• Overfitting
• Early stopping
• Drop out
• Normalisation
• Regularisation
• Hyperparameter Tuning of ANN

Convolutional Neural Network (CNN)

• Introduction to CNN
• Data augmentation
• RGB Scale/Gray Scale
• ANN vs CNN
• Convolutional Layer
• Strides and padding
• Pooling layer
• Max Pooling
• Avg Pooling
• Sum Pooling
• Implementation of CNN

Transfer Learning and Pretrain model

• Transfer learning
• Image classification using transfer learning
• Customised pre-train model
• Feature Extraction
• Fine tuning
• Steps need to follow in transfer learning
• Famous pre-train model available for image classification
• VGG16
• Inception (Google net)
• ResNet50
• Efficient Net
• Implementation of VGG16

Project on Deep Learning

NLP Introduction:

• ML vs DL vs NLP
• Basic terminology of NLP
• NLP Applications
• NLP pipeline 

Text Preprocessing 

• Tokenization - word tokenization, sentence tokenization , whitespace tokenizer, regex tokenization. 
• Stopwords removal 
• Language detection and translation with textblob, Google trans 
• Stemming, lemmatization 

Word Embedding 

• Count vectorizer 
• TF-IDF 
• Word2vec:
• CBow
• Skip-Gram 

NLP with Deep Learning

• Recurrent Neural Network (RNN)
• Long Short Term Memory (LSTM)
• Gated Recurrent Unit (GRU)

Real Time Project on NLP

 Transformers

• Encoder-Decoder Architecture
• Overview of Encoder-Decoder Models
• Applications in Translation, Summarization, and Beyond
• Attention Mechanism
• Scaled Dot-Product Attention
• Multi-Head Attention
• Role of Attention in Transformer Models
• Transformer Architecture
• Self-Attention Mechanism
• Positional Encoding
• Encoder-Decoder Framework
• Layer Composition
• Applications and Advantages of Transformers

Specialized Transformer Models

• BERT (Bidirectional Encoder Representations from Transformers)
• Understanding BERT's Architecture (Encoder-Only Model)
• Pre-training Objectives:
• Masked Language Modeling (MLM)
• Next Sentence Prediction (NSP)
• Fine-Tuning BERT for Downstream Tasks:
• Text Classification
• Sentiment Analysis
• Named Entity Recognition (NER)
• Question Answering
• Large Language Models (LLMs)
• GPT (Generative Pre-Trained Transformers)
• Fine-Tuning Techniques for LLMs
• Applications of LLMs:
• Text Generation
• Summarization
• Conversation Systems
• Vector Databases
• Basics of Vector Representations
• Popular Tools: FAISS, CosmosDB
• Integration of Vector Databases with Generative AI Models

Retrieval-Augmented Generation (RAG)

• Introduction to RAG
• RAG Architecture: Retriever and Generator
• Importance of RAG in Knowledge-Intensive Tasks
• Implementing RAG
• Tools for RAG: Hugging Face, FAISS, Pinecone
• Embedding Techniques and Challenges
• Applications of RAG in:
• Open-Domain QA
• Summarization

Practical Techniques and Applications

• Prompt Engineering
• Basics of Prompt Engineering
• Techniques to Optimize Prompts
• Real-World Examples and Best Practices
• Deployment of Generative AI Models:
• Platforms: AWS, Azure
• Use Cases: APIs, Chatbots
• Challenges and Best Practices
• Addressing Ethical Concerns in Generative AI
• Handling Bias and Robustness Issues

Hands-On Projects and Case Studies

• Sample Project names:
• Build a Retrieval-Augmented Question Answering (QA) System
• Create a Text Summarization Pipeline
• Deploy a Chatbot Using GPT Models

Introduction to Agentic AI

• Definition of Agentic AI
• Goal-driven AI systems
• Difference between GenAI, RAG, and Agentic AI
• From response generation to task execution
• Autonomous decision systems
• Practical business applications

Why Agentic AI

• Multi-step execution
• Goal completion
• Tool usage
• Dynamic decision making
• Failure handling
• Retry mechanism
• Workflow automation
• State tracking
• Controlled autonomy

Agentic System Fundamentals

• Goal definition
• Execution loop
• Planning capability
• Action selection
• Observation and feedback
• Replanning logic
• Termination conditions

Agent Control Loop

• Goal input
• Context state
• Plan creation
• Tool execution
• Result observation
• Loop continuation or completion

Tools and Function Calling

• Permission-based tools
• Schema-defined inputs
• LLM-driven tool selection
• Chained tool execution
• API integration
• Error recovery
• Secure invocation

Planning and Decision Making

• Task decomposition
• Step sequencing
• Conditional branching
• Iterative planning
• Adaptive execution
• Long-horizon reasoning

Memory in Agentic Systems

• Short-term state
• Long-term memory
• Context persistence
• Structured state storage
• Memory retrieval
• Context summarization

Agentic RAG

• Iterative retrieval
• Query refinement
• Multi-query execution
• Retrieval within agent loop
• Tool-assisted retrieval
• Cross-verification
• Multi-hop reasoning
• Confidence scoring
• Answer validation

Multi-Agent Systems

• Role-based agents
• Task delegation
• Agent communication
• Planner–executor model
• Hierarchical agents
• Parallel execution
• Collaboration strategies

Agent Architecture

• Orchestrator layer
• Agent runtime
• Tool layer
• Memory layer
• Observability layer
• Evaluation layer
• Monitoring and logging

Safety and Governance

• Access control
• Permission boundaries
• Approval checkpoints
• Audit logging
• Rate limiting
• Human-in-the-loop
• Compliance controls

Agent Frameworks

• Agentic AI vs frameworks
• LangGraph
• CrewAI
• AutoGen
• Framework selection strategy
• Stateful workflow design
• Multi-agent orchestration

Model Context Protocol (MCP)

• Agent-to-system control layer
• Intent separation
• MCP client
• MCP server
• Secure tool execution
• Auditable interactions
• Standardized protocol

Vibe Coding

• Introduction to Vibe Coding
• AI-assisted Software Development
• AI Coding Agents and Development Tools
• Code Generation, Testing, and Debugging
• Enterprise Use Cases and Best Practices 

Real-World Applications

• Enterprise workflow automation
• Compliance systems
• Incident response automation
• Knowledge assistants
• Transaction processing
• Autonomous reporting
• Business process automation

AWS for Data Science & Generative AI

• Understanding Cloud & Deployment Models
• Need of AWS in Data Science & GenAI
• Free tier account creation
• Introduction to AWS Console
• Regions and Availability Zones
• Billing dashboard & cost monitoring
• IAM users, roles and policies
• Understanding EC2 and compute services
• Creating and launching EC2 instance
• Security groups and key pairs
• Deploying ML / FastAPI / Flask project on EC2
• Introduction to AWS Lambda
• Introduction to API Gateway
• Amazon S3 – Introduction
• Creating buckets and uploading objects
• Bucket policies and access control
• Integrating S3 with ML workflows
• Versioning and lifecycle management
• Introduction to AWS RDS
• Introduction to AWS Glue (ETL overview)
• Introduction to Amazon Bedrock
• Foundation models overview
• Embeddings and vector search
• Building RAG pipeline on AWS
• Using OpenSearch as vector database
• Integrating S3 with Bedrock
• Deploying GenAI APIs
• Monitoring using CloudWatch
• Cost optimization and best practices

Azure for Data Science & Generative AI

• Understanding Azure fundamentals
• Creating Azure free account
• Azure Portal overview
• Resource groups and subscriptions
• Cost management and billing
• Azure Storage – Blob Storage & Data Lake
• Creating storage accounts
• Uploading and managing data
• Access control and managed identities
• Azure Virtual Machines overview
• Deploying ML / API project on Azure VM
• Introduction to Azure App Service
• Azure Machine Learning – Introduction
• Creating Azure ML workspace
• Building and training model
• ML pipelines and experiment tracking
• Model deployment (online & batch endpoints)
• Azure OpenAI – Introduction
• GPT models and embeddings
• Azure AI Search (Vector search)
• Building RAG pipeline on Azure
• Document Intelligence overview
• Deploying GenAI application on Azure
• Monitoring and logging
• Cost optimization strategies

Section 1: SQL (MySQL)

Topic 1: Database Fundamentals

• What is Database and SQL?
• Properties of Database and SQL
• Uses of Databases
• Types of SQL Commands

Topic 2: DDL and DML Commands

• DDL Commands and Implementation
• DML Commands and Implementation

Topic 3: SQL Clauses

• WHERE
• ORDER BY
• DISTINCT
• LIMIT
• GROUP BY
• HAVING

Topic 4: Advanced SQL

• Subqueries and Their Types
• Constraints in SQL
• NOT NULL
• UNIQUE
• PRIMARY KEY
• FOREIGN KEY
• Window Functions

Topic 5: SQL Joins

• INNER JOIN
• LEFT JOIN
• RIGHT JOIN
• SELF JOIN
• CROSS JOIN
• FULL JOIN

Section 2: MongoDB (NoSQL Database)

Topic 1: MongoDB Fundamentals

• Introduction to NoSQL Databases
• MongoDB Architecture
• SQL vs MongoDB
• Installing and Configuring MongoDB

Topic 2: Database and Collection Operations

• Create Database
• Create Collection
• Insert Documents
• Update Documents
• Delete Documents
• Query Documents
• Database Operations
• Collection Operations
• Indexing Basics 

Section 1: Flask

Topic 1: Flask Fundamentals

• What is Flask?
• Features of Flask
• Installing Flask and Setting Up the Environment
• Creating a Flask Application
• Application Structure

Topic 2: Routing and Request Handling

• Routing in Flask
• Path Parameters and Query Parameters
• Handling GET and POST Requests
• Request and Response Handling
• Working with JSON and Form Data

Topic 3: Templates and Static Files

• Using Jinja2 Templates
• Passing Data to Templates
• Managing Static Files (CSS, JS, Images)
• Dynamic URLs and URL Building
• URL For Function

Topic 4: REST API Development

• RESTful API Principles
• Creating REST APIs with Flask
• Working with JSON Data
• Status Codes and Responses
• Error and Exception Handling

Topic 5: Advanced Flask

• Middleware Concepts
• Authentication Basics
• File Upload and Download Handling

Section 2: FastAPI

Topic 1: FastAPI Fundamentals

• What is FastAPI?
• Features of FastAPI
• Installing FastAPI and Setting Up the Environment
• ASGI vs WSGI Overview
• Creating a FastAPI Application
• Running Applications with Uvicorn

Topic 2: Routing and Data Validation

• Application Structure and Routing
• Path Parameters and Query Parameters
• Request and Response Handling
• Handling JSON and Form Data
• Pydantic Models
• Status Codes and Response Models

Topic 3: REST API Development

• Creating REST APIs with FastAPI
• Automatic API Documentation (Swagger / ReDoc)
• Dependency Injection
• Error and Exception Handling

Topic 4: Advanced FastAPI

• Asynchronous Programming (async/await)
• Background Tasks
• Middleware Concepts
• Authentication Basics
• File Upload and Download Handling

NumPy(Numerical Python):

• Introduction to Numpy 
• Datatypes of ndarrays 
• Dealing with ndarrays, copies and views 
• Arithmetic operations, 
• Indexing , Slicing, splitting arrays 
• Shape manipulation 
• Stacking together different data 
• Stats using Numpy 
• Linear Algebric Operations 
• Logarithmic Functions 

Pandas (Data Analysis):

• DataFrame and Series 
• DataFrame operations 
• Data Slicing, indexing 
• DataFrame functions 
• Reading the files- csv, excel 
• Storing file in various formats 
• Useful DataFrame functions 
• Stats using pandas 
• Dealing with missing data 
• Operations over the data
• Sorting DataFrame Using Values and index 
• Groupy and Apply 
• Joins in Pandas

Visualisation Tools:

• Matplotlib 
• Seaborn 
• Plotly(3D Plotting)

Preprocessing and Model Building:

• Scipy 
• Stats models
• Sklearn 
• Nltk 
• Imblearn 
• Mlxtend