Module 24: LLM Applications

Chapter Overview

Large language models are reshaping virtually every industry. What began as research curiosities have become production tools that write code, analyze financial reports, assist doctors, detect cybersecurity threats, and power educational platforms used by millions. This module surveys the most impactful application domains for LLMs, providing both the technical depth needed to build these systems and the domain awareness required to deploy them responsibly.

The module opens with vibe-coding and AI-assisted software engineering, covering how tools like Cursor, Copilot, and Claude Code are transforming how developers write software. It then explores finance and trading applications, healthcare and biomedical AI, recommendation and search systems, cybersecurity applications, education and creative industries, and finally robotics and scientific discovery. Each section examines the unique challenges, architectures, evaluation methods, and regulatory considerations specific to its domain.

By the end of this module, you will understand how LLMs are applied across major industries, be able to design domain-specific LLM applications with appropriate safeguards, and appreciate the regulatory and ethical considerations that vary by sector.

Learning Objectives

Build AI-assisted coding workflows using code completion, agentic coding, and context engineering
Design financial NLP pipelines for sentiment analysis, report generation, and trading signals
Understand medical LLM applications and navigate HIPAA/FDA compliance requirements
Implement LLM-powered recommendation and conversational search systems
Apply LLMs to cybersecurity tasks including threat intelligence and vulnerability detection
Evaluate LLM applications in education, legal, and creative domains
Design embodied AI systems that use LLMs as planners for robots and software agents
Assess regulatory risks and ethical considerations across different application domains

Prerequisites

Module 09: LLM APIs (chat completions, function calling, structured outputs)
Module 10: Prompt Engineering (few-shot, chain-of-thought, domain-specific prompting)
Module 19: Retrieval-Augmented Generation (RAG pipelines, embedding search)
Module 21: AI Agents (tool use, planning, agentic architectures)
Module 22: Multi-Agent Systems (orchestration, collaboration patterns)
Domain-specific familiarity helpful but not required for individual sections

Chapter Overview

Learning Objectives

Prerequisites

Sections