Live Cohort Sessions

• •Before using ML
• •Building ML Pipelines
• •Model Serving and Deployment
• •Monitoring and Observability

We'll build a production-ready ML pipeline and deploy it to AWS. The lab will cover:

• •Data exploration and preprocessing
• •Model training and evaluation
• •Building a REST API for model serving using FastAPI
• •Deploying the model to AWS using Docker and ECS
• •Monitoring the deployed model using AWS CloudWatch

• •The Breakthrough - Transformers
• •LLMs
• •Prompting
• •LLMs can go wrong - Evaluation
• •Looking Ahead

We'll learn how to write effective prompts and evaluate LLMs on custom datasets. The lab will cover:

• •Principles and techniques for writing effective prompts
• •Preparing datasets for evaluation
• •Choosing a model and evaluation metrics
• •Evaluating LLMs on custom dataset and analyzing the results

• •The modern AI stack
• •LLM wrappers
• •Chatbots
• •RAG
• •Agentic systems
• •LLM fine-tuning

We'll learn how to fine-tune a small LLM on a custom dataset. The lab will cover:

• •Preparing the dataset for fine-tuning
• •Choosing a model and training parameters
• •Fine-tuning the model
• •Evaluating the fine-tuned model