LLM Adoption in the Enterprise

Book description

The emergence of generative AI marks a pivotal moment for product teams. Across industries, executives and product leads need to understand how they can use AI to propel their organizations forward with innovative and meaningful products.

Using large language models (LLMs) as an example, this concise guide illustrates how enterprises should approach the introduction of generative AI into their business models. From a practical understanding of the technology, to efficient and dynamic cross-functional product teams, to agile product lifecycles: packed with real-world examples, this report aims to equip thought leaders with a readily applicable framework for building with AI.

deepset's Isabelle Nguyen shares her experience from years of accompanying generative AI as it learned how to fly. Through her practical insight into the process of building with LLMs, you'll gain a rare understanding of the technology in action that is both pragmatic and visionary.

This report covers:

  • Product-oriented generative AI: Gain a practical understanding of LLMs and AI technology at large.
  • The product mindset: Learn about the importance of a product mindset and how to use it to effectively manage the product lifecycle.
  • Use cases and how to handle them: Discover how to map a tangible business use case to AI-driven solutions.
  • Cross-functional product teams: Find out about the incredible benefits that cross-functional teams offer through their diverse skill sets—and how to harness them through effective team leadership.
  • The AI team's toolkit: Examine the specific skills and knowledge that every AI team must have.

Isabelle Nguyen, technical content writer at deepset, is a linguist by training with a master's degree in NLP and machine learning. Her primary focus is demystifying the intricacies of generative AI and LLMs and making these complex topics understandable to a wide audience.

Table of contents

  1. Introduction
  2. 1. AI in the Enterprise
    1. AI in Context
      1. AI Today Is Mostly Machine Learning
      2. An Algorithm Trains a Model with Data
      3. A Trained Model Is Defined by Its Parameters
      4. There Are Many Different Kinds of Language Models
    2. So What Exactly Is an LLM?
    3. A Snapshot of Language Models in the Enterprise
    4. Four Sample LLM Applications
      1. Sophisticated Recommendations for Legal Professionals
      2. Conversational AI for Technical Documentation
      3. Automating the Collection of Information from Earnings Reports
      4. Condensing Political Discourse for News Consumers
    5. Pivoting Attention from Technology to Product
  3. 2. Building an AI Product
    1. What Is an AI Product?
    2. Considerations for Building with AI
      1. Defining a Use Case
      2. Understanding the Technology
      3. Assembling the Right Team
    3. From Product Ideation Toward Planning
  4. 3. The AI Product Development Lifecycle
    1. An Exemplary Case Study: Building a News Digest App
    2. Developing a Product Hypothesis
      1. Defining the Problem
      2. Identifying a Technical Solution
    3. Prototyping, Experimentation, and Evaluation
      1. Shipping a Prototype
    4. Refinement
      1. Back to the Drawing Board
      2. Another Round of Experiments, Refinement, and…More Experiments
  5. 4. The AI Team’s Toolkit
    1. Model Selection
    2. Prompt Engineering
    3. The Role of Data
      1. Training and Fine-Tuning
      2. Evaluation
      3. Retrieval Augmentation
      4. Best Practices for Data Curation
    4. Advanced Composable LLM Setups
      1. What Is Composability?
      2. Loops and Branches
    5. Putting the Pieces Together
  6. 5. From Pilot to Product
    1. Scalable Pilots
    2. Getting Ready for Production
    3. The Challenge of Monitoring LLMs
    4. Keep an Eye on AI Regulation
  7. Building Meaningful AI Products
  8. About the Author

Product information

  • Title: LLM Adoption in the Enterprise
  • Author(s): Isabelle Nguyen
  • Release date: March 2024
  • Publisher(s): O'Reilly Media, Inc.
  • ISBN: 9781098151645