Generative AI for Natural Language Processing

This program enables you to build and deploy cutting-edge Generative AI-driven solutions for various business problems. Gain conceptual and hands-on skills in applying Large Language Models (LLMs) and Generative AI techniques to text data.

Next cohort starts: To be announced

Enquire: +1 512 764 1278

16-week online program
15 mentored learning sessions
14 real-world case studies & 2 hands-on projects
GPU Labs Provided by Great Learning for Projects
Shareable e-Portfolio
Certificate from Great Learning

1,600 USD

ENROLL NOW

About The Program

In this 16-week online program, you will enhance your problem-solving skills using Generative AI. You'll dive deep into Large Language Models (LLMs), mastering scalable Prompt Engineering and performance optimization through fine-tuning techniques. Additionally, you'll gain expertise in building advanced RAG applications and intelligent Conversational Assistants with LangChain.
Note: Prerequisites for the program are Data Science and Machine Learning

What makes this program unique

What you’ll learn

Get hands-on experience with Large Language Models and Generative AI, from foundational Machine Learning to advanced Prompt Engineering for improved NLP performance.
Master practical applications and hands-on skills

Immerse yourself in 14 real-world case studies, projects like GenAI-Powered Support Ticket Resolution and Customer Review Classification & Summarization, and deploy a Generative AI-powered application.
Expert guidance from industry mentors

Gain practical knowledge in Generative AI from seasoned industry experts specializing in NLP solutions. Benefit from their guidance and support throughout the program.
Dedicated learning support

Get personalized assistance from Program Managers throughout your learning journey. Gain prompt query resolution through academic support and access to a project discussion forum and GL Community.

Skills You Will Gain

Prompt Engineering
Text Classification, Summarization & Generation
Transformer Embeddings for Search
Retrieval Augmented Generation (RAG)
Conversational AI with LangChain
Memory, Reasoning, and Querying Abilities
Optimization and Alignment for LLMs
Parameter-Efficient Fine-Tuning
Responsible AI

Tools You Will Learn

HuggingFace
LangChain
OpenAI
Llama-CPP
ChromaDB
Unsloth
LLama Guard
gte-large
Docker
Flask

What you will learn in this Program

Foundations of Generative AI

Understand the theoretical foundations of Generative AI and its applications.

Creating Conversational AI Assistants

Utilize the LangChain framework to create general-purpose AI Assistants.

Building Advanced RAG Applications

Develop advanced RAG applications that address and reduce the hallucination issue in LLMs.

Optimizing LLM Prompt Engineering

Craft effective prompts to interact with Large Language Models and obtain suitable outputs.

Developing Generative AI solutions for NLP

Create GenAI solutions for Text Classification, Text Summarization, and Generation.

Fine-Tuning for LLM Performance

Perform fine-tuning, QLoRA fine-tuning, and Alignment fine-tuning to improve LLM Performance.

Mitigating Security Risks

Implement best practices to mitigate security risks in Generative AI solutions.

Curriculum

Module 0 | Pre-work

Week 0 -Precursors to Generative NLP

This pre-work quasi-course is designed to help you brush up on the fundamental concepts and techniques in Machine Learning, Text Preprocessing, and Deep Learning utilized in Generative AI and Natural Language Processing. You will obtain an overview of the fundamental Python libraries and functions used in Machine Learning, the level of Text Preprocessing required for NLP, and key historical ideas from Deep Learning such as Word2Vec and RNNs that have been applied to Text Representation as well as Natural Language Understanding.

Topics Covered:

Machine Learning Refresher
Text Preprocessing for NLP

Module 1 | LLM Pathway to Generative AI

Week 1 - The Pathway to Generative AI

This week, you will gain your first understanding of the big picture behind Large Language Models and Generative AI. You will learn how AI has progressed from Discriminative Statistical & Probabilistic Techniques to Deep Learning methods and finally to Large Language Models and Generative AI, and the specific parts of this progression of NLP that will be covered in depth in the rest of this course. You will also learn about the framework of a typical Generative AI solution for NLP, the components and mechanisms involved in this solution, and how the conceptual knowledge required to understand these will be covered in this course.

Topics Covered

Historical Progression of NLP
LLMs and the Generative AI Timeline
Large Language Models and Generative AI
Components of a LangChain Generative AI Assistant
Solution Framework - Conversational AI Assistant

Sample Case Study: Demonstration: Sentiment Classification on IMDB dataset

Tools / Techniques: Python, Google Colab, HuggingFace, Llama-cpp

Week 2 - Transformers for Large Language Models

Transformers for LLMs is the first deep dive into the Transformer Neural Network, the key Deep Learning architecture behind Generative AI today. In this week, you will take the first steps towards opening up the black box and obtaining crucial insights into how Transformers and Self-Attention work under the hood. You will first be introduced to the original Transformer architecture, its Encoder & Decoder stages, the sequence of operations taking place, and some of the first-generation Transformer architectures introduced for NLP tasks. You will also understand the concept of Self-Attention in GPT-style Decoder-only Transformers, how Self-Attention represents the key innovation behind Transformers’ increased contextual understanding of Natural Language, and how it enables LLMs to provide the outputs they generate.
Topics Covered

High-Level Understanding of Transformer Architecture
Role of Encoder and Decoder
The Transformer Neural Network
Self-Attention in GPT-style Transformer LLMs

Sample Case Study: News Article Classification with LLMs

Tools / Techniques: Python, Google Colab, HuggingFace, Llama-CPP

Module 2 | LLM Foundations for Generative AI

Week 3 - Prompt Engineering Fundamentals

This week, we will take a step back from the core Transformer architecture and begin directly leveraging Transformer-based Large Language Models through external prompts. You will obtain a framework for understanding how these models convert your input prompt queries into sentence embeddings and utilize that to sample from an internal distribution and produce high-quality probabilistic outputs. You will be introduced to the field of Prompt Engineering - the primary port of call for interacting with LLMs and instructing them to produce outputs that are aligned with business objectives. You will utilize important Prompt Engineering techniques in both a No-Code proof-of-concept manner as well as integrate it into code-oriented applications.
Topics Covered:

Prompt Engineering for LLMs
Prompt Parameters vs Prompt Templates
No Code & Code-oriented Prompt Engineering
Zero-shot & Few-shot Prompting
Chain-of-Thought Prompting
Advanced Prompt Engineering Techniques

Sample Case Study: Customer Complaint Processing for E-commerce Platform Using Prompt Engineering

Tools / Techniques: Python, Google Colab, HuggingFace, Llama-CPP

Week 4 - Prompt Engineering Solutions for Generative NLP

The contents of this week will give you an opportunity to segue directly from the field of Prompt Engineering, towards actually utilizing LLM-based Prompt Engineering solutions for specific NLP tasks such as Text Classification, Text Summarization and Text Generation, and in doing so, observe how Generative AI has subsumed various earlier tasks in Discriminative and Generative NLP into just specific examples of its larger Generative objective. You will utilize Prompt Engineering to communicate important details regarding the business objectives of the task to the model, and in doing so, set the stage for the Conversational Assistant style solutions that the LLMs of Generative AI are most commonly used to develop.

Topics Covered

Text Classification, Text Summarization and Text Generation
Evaluation Metrics- Discriminative (Text-to-Label) & Generative (Text-to-Text) NLP
Workflow for Prompt Engineering-Based Generative NLP Solutions
BLEU, ROUGE, BERTScore
LLM-as-a-Judge: LLMs Evaluating Other LLMs
Generative NLP Solution Workflow

Sample Case Study: Sentiment Classification on IMDB Reviews

Tools / Techniques: Python, Google Colab, HuggingFace, Llama-CPP

Week 5 - Learning Break

Week 6 - Mid-term Project: Generative-AI Powered Support Ticket Resolution

In this week, you will embark on a hands-on project to delve into a business application of Prompt Engineering leveraging customer feedback for organizational growth and customer satisfaction. You will develop a Generative AI application using a Large Language Model to automate the classification and processing of support tickets.
Topics Covered

Hands-on Prompt Engineering techniques to extract information from and respond to customer responses
Workflow for JSON Data generation using Prompt Engineering-based Generative NLP solutions
Solution workflow for Generative Text Classification & Text Generation with LLMs
Applied Prompt Engineering
Prompt Evaluation & LLM Output Evaluation

Sample Case Study: Customer Complaint Classification and Summarization for a Finance Organization

Tools / Techniques: Python, Google Colab, HuggingFace, Llama-CPP

Module 3 | LLM Science and Engineering for Generative AI

Week 7 - AI Assistant Development with LangChain

In this week of the curriculum, you will be introduced to the idea of utilizing tools and agents to improve the functionality & utility of vanilla LLMs. You will be introduced to the LangChain framework to create solutions that essentially transition the LLM to becoming the manager of a solution framework, as opposed to being a sole individual contributor. You will also learn how to build custom classes / versions of these frameworks yourself, and how these AI assistant solutions can be augmented with custom tools, memory, human-like reasoning and the ability to query documents and data.
Topics Covered:

The LangChain Framework
Tools and Agents with LLMs
Augmenting LLM Capabilities with LangChain
Connecting LLMs with External Tools
Constructing Prototypical LLM Agents

Sample Case Study: Build a Data Science Agent that can use Python to crunch data from Natural Language inputs

Tools / Techniques: Python, Google Colab, Langchain, HuggingFace

Week 8 - Semantic Search and Retrieval-Augmented Generation

In the content for this week, you will obtain an overview of Semantic Search - a key application of the increased contextual Natural Language Understanding that Transformers display over other text embedding schemes. You will be introduced to how Semantic Search leverages Transformer Sentence Embeddings to return the most relevant results per your search query, how this property is utilized in Generative AI solutions that require breaking down large text corpuses into chunks and storing them in a vector database, and how this idea can be applied to increase the capabilities of a Conversational AI assistant. You will then transition to Retrieval-Augmented Generation (RAG), a key solution space in LLMs and Generative AI that uses Semantic Search to help LLMs answer factual and contextual queries by retrieving from vector databases. You will finally learn about evaluation metrics to understand the performance of your RAG solutions, and how the vanilla RAG process can be tweaked in multiple ways to create more performant RAG applications.
Topics Covered

Semantic Search with Sentence Embeddings
The RAG Workflow for Contextual Prompts
Transformer Embeddings
Semantic Search
Retrieval-Augmented Generation (RAG)

Sample Case Study: RAG QnA Chatbot on Annual Reports

Tools / Techniques: Python, Google Colab, ChromaDB, Llama-CPP, gte-large

Week 9 - Optimization, Alignment and Fine-tuning for LLMs - Part I

In this week, we will deep dive into the idea of Fine-tuning, how it represents a deeper level of involvement than Prompt Engineering or RAG, and how we can apply Parameter-Efficient Fine-tuning to Large Language Models. LLMs are otherwise prohibitively compute-costly to fine-tune using regular methods, making PEFT a necessity for most fine-tuning requirements. We will primarily focus on the Quantized Low-Rank Adaptation (QLoRA) method and use that to fine-tune open-source LLMs for task-specific requirements.
Topics Covered:

Fine-Tuning for Large Language Models
Necessity, benefits & risks with Fine-Tuning
Parameter-Efficient Fine-Tuning (PEFT)
Quantized Low-Rank Adaptation (QLoRA)

Sample Case Study: Classify Stackflow questions into high and low priority using a QLoRA finetuned model, prompt engineering, and a fine-tuned model

Tools / Techniques: Python, Google Colab, BERT finetuning, Unsloth, Llama-CPP, HuggingFace

Week 10 - Optimization, Alignment and Fine-tuning for LLMs - Part II

In this week of the program, we deep dive into the overall GPT-style Assistant Training Pipeline, and examine the end-to-end process of training, fine-tuning and aligning LLMs with human preferences. Through this, we will go over the difference between the Supervised Fine-tuning Stage and the Human Alignment Stage, and we will examine various strategies being used by the open-source LLM community to optimize & fine-tune high-quality performant LLMs at small sizes. This will provide complete clarity into the options available & the process required for training & fine-tuning custom LLMs from the ground up, a valuable set of skills for organizations looking to lessen / eliminate their reliance on proprietary, closed-source LLMs.
Topics Covered:

The GPT-Style Assistant Training Pipeline
Training & Fine-Tuning Custom LLMs
Direct Preference Optimization (DPO)
LLM Inference Optimizations

Sample Case Study: AI-powered coding Assistant using Alignment Fine-Tuning (partial training)

Tools / Techniques: Python, Google Colab

Week 11 - Learning Break

Week 12 - Project-2: Customer Review Classification & Summarization

In the final project of our Generative AI for NLP course, students will engage in an advanced exploration of Generative AI applications, focusing on two cutting-edge topics: Sentiment and Aspect-Based Classification and Fine-tuning LLMs for Summarization Tasks. This project is designed to build upon the skills developed during the Mid-term project, expanding the application of Generative AI in NLP to more nuanced tasks.
Topics Covered:

Solution Workflow for Generative Text Classification & Text Generation with LLMs
Applied Prompt Engineering
Fine-tuning LLMs: Practical experience in fine-tuning generative models for specific NLP tasks, focusing on summarization
Prompt Evaluation & LLM Output Evaluation

Sample Case Study: Fine-tuning and prompt engineering for Aspect Based Sentiment Analysis on E-commerce Reviews

Tools / Techniques: Python, Google Colab, HuggingFace, Llama-cpp, Unsloth

Module 4 | Advanced Generative AI

Week 13 - Designing Custom Agentic Workflows

In this week of the curriculum, we will go deeper into the concepts of agents. We will learn Function Calling and learn how to override off-the-shelf agent classes from Langchain to create advanced, custom agents.

Topics Covered:

Function Calling
Overriding agent classes in Langchain to implement faster agents
Construct complex agent workflows

Sample Case Study: Multi-step Finance Agent to automate Investment decisions

Tools / Techniques: Python, Google Colab, OpenAI API, Langchain, HuggingFace

Week 14 - Advanced RAG Techniques

In this week, you will enhance each step of the Vanilla-RAG workflow (From week-8) to create an advanced RAG application. We will parse and index multi-modal data (text, image, and tables), apply advanced search and retrieval techniques like query expansion, re-ranking, Hypothetical Questions, etc, and then finally learn to evaluate the RAG application using the RAGAS Framework.

Topics Covered

Parse and Index Multi-Modal Data
Re-Ranking
Query Expansion
RAGAS

Sample Case Study: AI Research Assistant - Answer scientific questions using research papers

Tools / Techniques: Python, Google Colab, ChromaDB, Llama-CPP, gte-large

Week 15 - Building Secure and Responsible AI Solutions

This week, we will explore the key ethical and legal challenges surrounding Generative AI solutions, focusing on the top risks involved in deploying such applications. We will dive into the CIA Triad (Confidentiality, Integrity, and Availability) and learn how to address major security concerns by building guardrails using Lamaguard and block lists.

Topics Covered:

CIA Triad
Top Risks for Gen-AI Solutions - (Prompt Injection, Jail-Break, Sensitive Information Disclosure, etc.)
Guard-Railing for LLM Applications

Sample Case Study: Build Multiple guardrails to safeguard a consumer-facing LLM Application

Tools / Techniques: Python, Google Colab, Lamaguard

Week 16 - Deploying a Generative AI-Powered App

This week, we will dive deep into the fundamental concepts of deploying a Generative AI-powered application. You will learn the essentials of REST API and building API inference using Flask. Additionally, we'll cover the basics of containerization and deploy the Generative AI application, enabling you to perform batch inference efficiently.

Topics Covered:

REST API
Containerization
Batch Inference

Sample Case Study: Deploy the RAG agent we built in week 14

Tools / Techniques: Python, Google Colab, Docker, REST API

Earn Your Certificate

Enhance your resume with a certificate in Generative AI for Natural Language Processing from Great Learning and share it with your professional network.

Get to Know Your Faculty & Mentors

Learn from highly skilled professionals in the field of Data Science who have engineered NLP solutions across industry verticals and have real-world, hands-on work experience.

Dr. Abhinanda Sarkar

Academic Director - Data Science & Machine Learning

Dr. Pavankumar Gurazada

Senior Faculty & Director Academics, Great Learning

Prof. Mukesh Rao

Director, Academics, Great Learning

Georg Huettenegger

Mentor, Great Learning

Omar Attia

Senior Machine Learning Engineer

Anuj Saini

AI Research Scholar,Université de Montréal

Davood Wadi

AI Research Scientist, intelChain

Fred Premji

Principal AI/ML Engineer, OPTMAL

Hassan Saidinejad

Data Scientist, Intact

Success Stories

Our learners share their program experiences

"I am very pleased with the program as it includes up-to-date content on Generative AI and NLP techniques, with material and projects centered on current AI methodologies and trends. We had supportive advisors who provided meaningful insights alongside exceptional mentors available to help you reach your professional goals."

Stefani Colossi

Machine Learning Consultant, Business Owner

"It is a well-structured program with plenty of real-life business use cases, demonstrating processes and methodologies using the latest LLMs. It has improved my expertise in AI and Prompt Engineering and has also helped me a lot at work."

Vladimir Uzunov

Associate Manager, Accenture

"The course exceeded my expectations with highly relevant content, comprehensive coverage of all topics with detailed explanations, and engaging video lectures. The instructors, who were industry professionals, demonstrated expertise in AI, enriching the learning experience by sharing valuable insights from their real-world experiences."

Sandeep Goswami

Software Engineering Manager, Citizens Bank

"Great Learning structured Training videos and reading documents in such a way that I was able to understand fundamentals, solve quizzes and complete project assignments. The best part of whole program is weekly mentored learning sessions conducted by industry expert. The projects boosted my confidence level that I am applying all these projects learning into my real data science projects."

Rizaldy Nuque

Tech Innovation Director, Accenture

Still have queries?
Contact Us

Please fill in the form and a Program Advisor will reach out to you. You can also reach out to us at learningforlife@mygreatlearning.com or +1 512 764 1278.

Application Closes 24^th Dec 2024

Download Brochure

Check out the program and fee details in our brochure

Name

Mobile Number

By submitting this form, you consent to our Terms of Use & Privacy Policy and to be contacted by us via Email/Call/Whatsapp/SMS.

1,600 USD

PG Program in Artificial Intelligence and Machine Learning

Applied Data Science Program

No Code AI and Machine Learning: Building Data Science Solutions

PG Program in Data Science and Business Analytics

MIT Data Science and Machine Learning Program

PG Program in Cloud Computing

Post Graduate Program in User Experience Design

Certificate Program in AI Business Strategy

No Code AI and Machine Learning: Building Data Science Solutions

PG Program in Artificial Intelligence and Machine Learning

Applied Data Science Program

PG Program in Artificial Intelligence for Leaders

Certificate Program in Applied Generative AI

MIT Data Science and Machine Learning Program

Microsoft AI Professional Program (AI to OpenAI)

MS in Data Science Programme

Applied Data Science Program

PG Program in Data Science and Business Analytics

MIT Data Science and Machine Learning Program

Data Analytics Essentials

MS in Data Science Programme

Data Analytics and Power BI Bootcamp

PL-300 - Microsoft Power BI Data Analyst Certification Training

Post-Graduate Program in Data science & Analytics

Executive PG Program in Management

PGP in Strategic Digital Marketing

AZ-104 - Microsoft Azure Administrator Training Course

AWS Solutions Architect Associate Certification Training Program

PL-300 - Microsoft Power BI Data Analyst Certification Training

Microsoft AI Professional Program (AI to OpenAI)

PG Program in Cloud Computing

AZ-104 - Microsoft Azure Administrator Training Course

Post Graduate Program in Cyber Security

CompTIA Security+ Bootcamp

Professional Certificate in Full Stack Software Development: Building Scalable Cloud Applications

Executive PG Program in Management

PGP in Strategic Digital Marketing

Post Graduate Program in User Experience Design

MS in Data Science Programme

About The Program

What makes this program unique

Skills You Will Gain

Tools You Will Learn

What you will learn in this Program

Foundations of Generative AI

Creating Conversational AI Assistants

Building Advanced RAG Applications

Optimizing LLM Prompt Engineering

Developing Generative AI solutions for NLP

Fine-Tuning for LLM Performance

Mitigating Security Risks

Curriculum

Module 0 | Pre-work

Module 1 | LLM Pathway to Generative AI

Module 2 | LLM Foundations for Generative AI

Module 3 | LLM Science and Engineering for Generative AI

Module 4 | Advanced Generative AI

Earn Your Certificate

Get to Know Your Faculty & Mentors

Success Stories

Still have queries? Contact Us

Download Brochure

Oops!! Something went wrong, Please try again.

Form submitted successfully

Hey there! Welcome back.

You are already registered. Please login instead.

Forgot your password? No problem.

You are already registered. Please login instead.

Login

Forgot Password?

Still have queries?
Contact Us