Generative AI Training in Ameerpet Hyderabad

Module 1: Python Foundations & ML Basics

1. Python Programming Essentials

• Data types: strings, lists, tuples, dictionaries

• Control flow & functions

• OOP concepts (overview)

2. Advanced Python Data Structures

• List & dictionary comprehensions

• Sets & decorators (optional)

3. Introduction to Machine Learning

• Scikit-learn overview

• Train-test split

• Linear & Logistic Regression

4. Evaluation Metrics

• Accuracy, Precision, Recall, F1-score

• Confusion matrix visualization

Hands-On Labs

• Build a classifier using Iris dataset

• CSV parsing & summary statistics

• GitHub version control practice

Outcome: Strong Python foundation, ML workflow understanding, Git proficiency

Module 2: Data Analysis & ML Workflow

1. Data Handling & EDA

• NumPy, Pandas

• Matplotlib & Seaborn

• Data ingestion & cleaning

• Feature engineering

2. ML Workflow Best Practices

• Training vs validation

• Model evaluation & metrics

• Reproducibility

Hands-On Lab

• Titanic survival prediction project

• End-to-end ML workflow implementation

Module 3: Foundations of AI & Deep Learning

1. Neural Network Essentials

• AI vs ML vs Deep Learning vs Generative AI

• Perceptron model

• Forward & Backward propagation

• Activation functions: ReLU, Sigmoid

2. Deep Learning Frameworks

• PyTorch or TensorFlow basics

• Tensors, training loops, model definition

Hands-On Lab

• Train a CNN on MNIST or CIFAR-10

Module 4: CNNs & RNNs (Optional Advanced Basics)

1. Convolutional Neural Networks (CNNs)

• Image classification concepts

• CNN architectures

2. Recurrent Neural Networks (RNNs)

• Sequence data basics

• LSTM overview (time permitting)

Module 5: Generative AI & Large Language Models (LLMs)

1. Transformer Architecture

• Self-attention & Multi-head attention

• Positional embeddings

• Encoder–Decoder vs Decoder-only (GPT)

2. LLM Ecosystem

• OpenAI GPT (3.5, 4)

• Google Gemini, Meta LLaMA

• Tokenization: BPE, WordPiece

• Embeddings & prompt engineering

3. Fine-Tuning vs Prompt Engineering

• Few-shot & chain-of-thought prompting

• LoRA, Adapters, Parameter-efficient tuning

Hands-On Labs

• Prompt engineering with GPT or local LLaMA

• Intro to fine-tuning open-source LLMs

Module 6: Retrieval-Augmented Generation (RAG)

1. RAG Fundamentals

• Hallucination reduction

• RAG pipeline architecture

2. Vector Databases & Embeddings

• Pinecone, Chroma, Weaviate, Milvus

• OpenAI embeddings, Sentence Transformers

3. Implementing RAG Pipelines

• Document chunking & indexing

• Query retrieval & generation

• LangChain-based Q&A systems

Hands-On Lab

• Build a RAG-based Q&A system

• Benchmark retrieval precision & recall

Module 7: Agentic AI Workflows

1. AI Agents Concepts

• Autonomy & planning

• Multi-step reasoning

2. Agent Frameworks

• LangChain Agents

• Semantic Kernel / Crew AI (overview)

3. Tool & API Orchestration

• External APIs (weather, stock, DBs)

• Chain-of-thought prompting

Hands-On Lab

• Build an AI agent calling external APIs

• Skill-based agent workflows for business automation

Module 8: Advanced Fine-Tuning, Benchmarking & Scalability

1. Advanced Fine-Tuning

• LoRA, QLoRA, Adapters

• Domain-specific dataset creation

2. Benchmarking & Performance

• BLEU, ROUGE, Perplexity, FID

• Throughput, latency, memory usage

• Multi-GPU & distributed training (concepts)

3. Bias, Robustness & Reliability

• Bias evaluation

• Robustness testing

Hands-On Lab

• Domain-specific LLM fine-tuning

• Performance comparison vs baseline

Module 9: MLOps, AIOps & Production Deployment

1. Cloud & AIOps Pipelines

• CI/CD with GitHub Actions, Jenkins

• Automated deployment pipelines

2. Containerization & Serving

• Docker fundamentals

• Kubernetes vs Serverless

• FastAPI / Flask model serving

3. Observability & Monitoring

• Logging & metrics

• Drift detection

• Monitoring dashboards

Hands-On Lab

• Deploy RAG or AI Agent app to AWS / Azure / GCP

• Enable logs & performance monitoring

Module 10: FastAPI & Cloud Deployment

1. FastAPI Essentials

• REST APIs, async I/O

• CRUD operations

• Pydantic validation

• OAuth2 & JWT (overview)

2. Testing & Packaging

• Pytest

• Dockerfile creation

3. Multi-Cloud Deployment

• AWS ECS / Elastic Beanstalk

• Azure Web App for Containers

• GCP Cloud Run

Hands-On Labs

• Build & deploy FastAPI apps across clouds

• Integrate ML inference endpoints

Capstone Workshop 

1. Project Scoping

• Business or industry-focused problem

• Define performance & scalability goals

2. End-to-End Implementation

• Data prep & fine-tuning

• RAG or Agentic AI integration

• Benchmarking & monitoring

3. Responsible AI

• Bias & fairness checks

• Privacy & compliance (GDPR, HIPAA)

• Prompt security testing

4. Career Preparation

• Resume & LinkedIn optimization

• Interview readiness (ML + GenAI system design)

• GitHub & Hugging Face portfolio

Final Outcome

• Portfolio-ready Generative AI project

• Deployed ML/GenAI application on cloud