Document processing is usually the first, necessary step to building large-scale natural language processing (NLP) and AI systems.

This course provides an understanding of the common challenges and solutions for document processing in modern enterprises including optical character recognition (OCRing), table extraction, and how to handle form retrieved data.

Training to enable the design and optimization of prompts to guide AI models, particularly LLMs, towards generating the high quality responses.

Levels of training including:

1. Basics prompting Introduction and fundamentals.  

2. Advanced Prompting using Custom GPTs

3. Professional Prompting using Agentic Workflows and integrated tools

4. Expert Prompting using vibe coding with agents

RAG systems enable the building powerful document search engines and chatbots that can output accurate, context-aware answers using enterprise data.

This course delivers a comprehensive understanding of how to systematically design, build, optimize and operate RAG systems and chatbots.

Agentic AI is an architecture style that takes automation of existing work processes to the next level. "Agents" are software processes with a predefined level of autonomy, that allows them to analyze and react accordingly to make autonomous decisions within closely defined boundaries.

The Levels of Training Include:

1. Hands-On Agents (No-Code / Low-Code): Building and using agents without programming skills 

2. Agents Deep Dive (Developer Track): Advanced agent design, orchestration, and programming-based customization 

MLOps are a core function of Machine Learning engineering, focused on streamlining the process of taking machine learning models to production.

This course introduces the principles and available tools for creating scalable and secure ML and AI systems. Covering topics including:

1. Identification and education on the four key AI/ML risk categories: 
   Unintended failure

   Irresponsible use

   Intentional attacks

   Weaponization of AI/ML

2. Adversarial threat modeling and risk awareness training 

3. Understanding attack vectors such as poisoned datasets, backdoors, and model extraction 

4. Risk-informed decision support for responsible AI development 

Pharmaceutical-specific controls for AI are essential to ensure patient safety, data integrity, and regulatory compliance (GxP) in an industry with low tolerance for error. These controls are shifting from static, paper-based validation to dynamic, data-centric governance frameworks that manage AI throughout its lifecycle. Topics in this course incluce:

1. GMP considerations for data and AI workflows

2. Patient data safety and compliance 

3. Database-level security basics and access management