Building LLM Powered Applications

Packt via Coursera

Go to class Write review

Details

Go to class

Provider

Coursera
Pricing

Paid Course
Languages

English
Certificate

Paid Certificate Available
Duration & workload

19 hours 57 minutes
Sessions

Self-Paced
Level

Beginner
Subtitles

English

Found in

Overview

Google, IBM & Meta Certificates — All 10,000+ Courses at 40% Off

One annual plan covers every course and certificate on Coursera. 40% off for a limited time.

Get Full Access

This course provides a comprehensive introduction to building intelligent applications powered by large language models (LLMs). You'll explore foundational LLM concepts, architectural frameworks, and practical applications in real-world scenarios. By using leading LLM toolkits and frameworks, you'll gain hands-on experience in creating intelligent agents capable of handling both structured and unstructured data. The course emphasizes the integration of LangChain for orchestrating complex AI workflows and covers prompt engineering techniques essential for customizing and optimizing LLMs. What sets this course apart is its blend of theoretical learning and practical implementation, making it an ideal resource for those looking to implement LLMs in real-world applications. It ensures you can build LLM-powered applications from scratch while navigating the challenges of real-world scenarios, including ethical considerations. This course is suitable for software engineers, data scientists, and researchers who are keen on understanding the applied aspects of generative AI. No prior experience with LLMs is required, but a strong understanding of machine learning concepts will enhance your learning experience. Based on the book, Building LLM Powered Applications, by Valentina Alto.

Syllabus

Introduction to Large Language Models

In this section, we introduce Large Language Models (LLMs), discuss their role in generative AI, compare LLM architectures with classical machine learning, and explain the distinction between base and fine-tuned LLMs for real-world applications.

LLMs for AI-Powered Applications

In this section, we examine how large language models (LLMs) are transforming software development, explore the architecture of copilot systems, and evaluate AI orchestrator frameworks for embedding LLMs in real-world applications.

Choosing an LLM for Your Application

In this section, we examine the criteria for selecting large language models (LLMs), comparing architectures, performance, costs, and real-world trade-offs to optimize application integration and responsible use.

Prompt Engineering

In this section, we introduce prompt engineering techniques to create effective prompts that guide large language model behavior and help reduce bias and hallucinations.

Embedding LLMs Within Your Applications

In this section, we demonstrate how to embed large language models (LLMs) in applications using LangChain, integrate Hugging Face models, and leverage frameworks for enhanced conversational user experiences.

Building Conversational Applications

In this section, we build LLM-based conversational applications using LangChain, adding memory, non-parametric knowledge, and tools, while developing a Streamlit front-end for rapid prototyping and practical deployment.

Search and Recommendation Engines with LLMs

In this section, we examine how large language models (LLMs) modernize recommendation systems, discuss traditional and LLM-powered techniques, and implement practical applications using LangChain and Streamlit for interactive user experiences.

Using LLMs with Structured Data

In this section, we demonstrate how to integrate large language models (LLMs) with relational databases, enabling natural language interfaces to tabular data and combining structured with unstructured sources for practical applications.

Working with Code

In this section, we explore how Large Language Models (LLMs) support code generation, understanding, and algorithm emulation, enabling the development of natural language-driven programming tools and code-based applications.

Building Multimodal Applications with LLMs

In this section, we learn to build adaptive multimodal agents by integrating language, image, and audio models using LangChain and Azure AI, enabling robust, practical AI workflows and applications.

Fine-Tuning Large Language Models

In this section, we examine the theory and practical steps for fine-tuning large language models (LLMs), covering data preparation, domain-specific taxonomy, and implementation using Python and Hugging Face for specialized NLP applications.

Responsible AI

In this section, we examine Responsible AI practices for mitigating risks and biases in large language model (LLM) applications, exploring architectural strategies and key regulatory requirements to ensure safer AI deployment.

Emerging Trends and Innovations

In this section, we examine recent innovations in large language models (LLMs) and generative AI, explore enterprise adoption, and discuss applications such as GPT-4V(ision), AutoGen, and small language models for future-ready development.