Build Intelligent Apps with Generative AI

• Introduction to Python:variables,datatypes,control structures

• Functions, modules, error handling

• Numpy and Pandas basics for data manipulation

• Getting started with Streamlit

• Using widgets, layout, and components

• Creating interactive dashboards and LLM UI mockups

• Understanding supervised and unsupervised learning

• Scikit-learn model building: Linear & Logistic Regression

• Train-test split, cross-validation

• Hyperparameter tuning and evaluation metrics

• Use case: spam detection / customer churn

• Neural network basics: perceptron, multi layer networks

• Introduction to PyTorch tensors and autograd

• Building simple neural networks using nn.Module

• Training loop with loss functions and optimizers

• GPU acceleration and model evaluation

• Text: cleaning, stemming, lemmatization

• Vectorization using Bag of Words, TF-IDF, Word2Vec

• Sequence models: RNN, LSTM concepts

• NER, sentiment analysis hands-on

• Limitations of traditional NLP techniques

• Transformer architecture (self-attention,encoder-decoder)

• BERT, GPT-2 overview and use cases

• Using Hugging Face’s transformers pipeline

• Downloading and using pre-trained models

• Hands-on: sentiment classifier, text summarizer

• LangChain architecture:Chains,Agents, Memory

• Building LLM chains using prompts and tools

• Creating a conversational bot with LangChain

• Integration with OpenAI API

• LangChain toolkits: calculator, Python REPL

• Understanding RAG and how it enhances LLM’s

• Vector databases: FAISS, Pinecone basics

• Document loading and chunking for retrieval

• Creating context-aware Q&A systems

• Build a private chatbot using LangChain + RAG

• Prompt engineering patterns and anti-patterns

• Using Streamlit Cloud and Hugging Face Spaces for deployment

• Monitoring LLM behavior and optimization techniques

• Caching, truncation, rate limits and usage metrics

• Addressing hallucinations, bias, and safety in LLMs

• Plan and design a complete GenAI solution

• Possible projects: Resume screener, PDF-based chatbot, AI tutor

• Incorporate Streamlit + LangChain + RAG

• Model evaluation and improvement

• Presentation and deployment