• LLM Overview
• Evolution of LLMs
• Capabilities and Limitations of LLMs
• Applications and use cases of LLMs

• Tokenization, Vectors and Embeddings
• Attention Mechanism and its variants
• Introduction to Transformer Architecture
• Creating Custom Language Models
• Transfer Learning in NLP
• Evaluation Metrics for LLMs: BLEU, ROUGE, Perplexity
• Introduction to Hugging Face Transformers library
• Overview of llama2 and Gemma
• Fine Tuning Gemma Model

• Overview of popular LLMs: GPT-3/4, BERT, T5
• Fine-tuning pre-trained models for specific tasks
• BERT and its variants: RoBERTa, DistilBERT
• GPT and its applications in text generation
• Exploring other models: T5, XLNet, ELECTRA
• Building conversational agents and chatbots
• Creative applications: text generation, storytelling
• Ethical considerations and bias mitigation in LLMs

• Understanding Computer Vision
• CNN from Scratch
• CNN using Tensorflow

• Basics of audio signal processing
• Feature extraction: MFCCs, Spectrograms
• Audio classification and speech recognition
• Basics of video signal processing
• Frame extraction and video feature analysis analysis

• LangChain – Langchain for Conversational AI Applications
• LangChain – Deploying
• Language Model APIs with Langchain
• LangChain – Langchain for RAG Workflows
• Ollama – Overview of Ollama for conversational AI
• Ollama – Developing and deploying conversational agents with Ollama

• Overview of Text Classification Model
• Bert text classification
• Data preparation and preprocessing
• Text Generation Model

• Overview of Text Generation Model
• Evaluation and fine-tuning
• Evaluation and fine-tuning

• Overview of Designing a conversational agent architecture
• Conversational Agent
• Conversational agent using openAI
• Conversational Agent using LangChain
• Conversational Agent using Ollama
• Conversational Agent using HuggingFace

• Deploying LLMs with Flask and FastAPI
• Introduction to Docker for containerization
• Deploying LLMs on cloud platforms (AWS)
• Introduction to MLOps concepts and practices
• Continuous Integration and Continuous Deployment
• Monitoring model performance in production

Followed by the certification session, an exam will be conducted for a total of 100 marks.

You need to acquire 60+ marks to clear the exam.

If you fail to acquire 60+marks, you can retake the exam after one day.

The maximum number of retakes will be three.

If you fail to acquire 60+ marks even after three attempts, then you need to contact us to get assistance for clearing the exam.