Andreea is a data scientist with over 7 years of experience in demystifying AI and Data Science concepts for anyone keen on working in this exciting field using cutting-edge technology. Having obtained a Master’s Degree in Quantitative Economics and Econometrics from Lumière Lyon 2 University, she enjoys integrating machine learning principles with real-world applications. Andreea’s passion lies in developing engaging training programs and ensuring an optimal customer education journey. As she frequently likes to remark, “AI is essentially Economics turbocharged by data, with a sprinkle of innovation.”
Large Language Models (LLMs) - Level 2
Udemy Course Rating:4.3★★★★★★★★★★
- Generative AI
- H2O Cascade
- Last updated 01/2025
- English
- Intermediate
- 1h 31m · 7 lessons
Created by:
Share course
About this course
Build on your foundation in Large Language Models (LLMs) with practical techniques for fine-tuning, optimizing, and deploying NLP models.
This course focuses on hands-on experience with H2O LLM Studio, guiding you through model customization, experiment monitoring, and deployment.
Explore advanced topics like quantization and LoRA to make your models efficient and production-ready. Designed for learners with prior exposure to LLMs, this course helps deepen your understanding and awards you a certification upon completion.
What you'll learn
Skills you will take away
Importance of Clean Data for NLP
Understand why data quality matters and how well-prepared datasets drive reliable, fair, and high-performing language models.
Hands-On with LLM Studio
Navigate the no-code interface to configure, launch, monitor, and compare fine-tuning experiments with ease.
Fine-Tuning Techniques
Learn when and why to fine-tune LLMs, and how specialization, transfer learning, and task-specific data shape model performance.
Choosing the Right Backbone
Explore how to select pre-trained model architectures based on task, domain, data size, and hardware constraints.
Model Compression Essentials
Apply advanced techniques to reduce model size and speed up inference without sacrificing performance.
Deploying to Hugging Face
Export your fine-tuned model and share it with the broader AI community through one-click deployment.
Course playlist on YouTube
Open the playlist (opens in a new tab)7 lessons, start to finish
Pick a lesson to play it here. Watch in order, then take the quiz.
In this video you are going to maximize the language model performance with Data Preparation! Explore key functions like data augmentation, text cleaning, profanity checks, and more. Plus, learn how to structure unstructured data into Q&A format for efficient model training. Disclaimer: Please note that certain content displayed has been created utilizing both our H2OGPT and Open AI's GPT-3.5 platforms. This is done to demonstrate the versatility of these tools in recognizing textual patterns and resolving complexities in language model (LLM) applications. PS: This video is a part of a series published on our LLM Learning Path Playlist, that you can check out here: https://youtube.com/playlist?list=PLNtMya54qvOHQHDpUDtZytwEV2Miali9l&si=lIXmA0hqGZhftSZe
Quiz Me if You Can!
Resources
Additional course resources and access links
Access Course Slides (opens in a new tab)Via GitHub RepositoryWatch the Course (opens in a new tab)Via YouTube1 hour 48 minsFor all Learners (opens in a new tab)Get Certified via Udemy · free with code OCT-FREE20261 hour 58 minsFor all Learners (opens in a new tab)Get Certified via Coursera1 hour 58 mins
Your instructors
Taught by the people who build and deploy it
Continue the learning path
View this path