Chapter 18: Where to go from here
The introduction marks the reader’s progress from basic programming to a local AI chatbot and frames choosing a next project as the first step forward.
After this chapter you will be able to
- Choosing a next project. Explain what first step the introduction recommends for continuing after the book, regardless of whether the aim is personal use, workplace use, or an AI engineering career.
- Layered AI development skills. Describe how the book’s skills build from working with local models to building AI applications and understanding more advanced approaches. Give concrete examples from the recap without explaining their implementation.
- Practical independence with AI. Explain the practical shift the section claims: from using other people’s AI tools to building and evaluating tools for yourself or others. Describe examples of the control this can provide over deployment, sensitive data, and costs.
- Transferable local AI foundation. Explain how the working local AI environment and development skills gained in the book prepare the reader to follow new tutorials and extend their projects without relearning the basics.
- One Project at a Time. Explain why the section recommends choosing a project that solves a real problem and completing it before exploring other intermediate topics.
- Independent advanced learning. Explain how learning changes at the advanced level and what remains the same about the learning process.
- Enduring AI Building Patterns. Explain why the skills developed in the book remain useful as models, frameworks, and APIs change. Describe the recurring work of connecting a model to an interface, handling responses and state, and debugging errors.
- Practice Sharing And Curiosity. Describe the section’s three principles for continuing to learn: start building, share what you learn, and stay curious. Explain how practice and teaching help when progress feels difficult or tools change.
What it builds on
- The complete voice chat application (chapter 10) · The completed voice chat application is an example of the local AI application summarized here.
- Local AI data flow (chapter 14) · Local AI data flow provides context for keeping sensitive data on your own hardware.
- Agent cost analysis (chapter 17) · Agent cost analysis provides context for comparing local deployments with proprietary APIs.
- The complete virtual environment workflow (chapter 6) · The virtual environment workflow supports starting Python projects and installing dependencies.
- The complete chat application data flow (chapter 11) · The chat application data flow provides an example of the recurring patterns.
- Reading error messages effectively (chapter 15) · Reading error messages is one of the transferable debugging skills.
- Verifying offline chatbot operation: The airplane mode test (chapter 14) · Offline chatbot operation establishes the local option referenced here.
Section by section
Introduction
The introduction marks the reader’s progress from basic programming to a local AI chatbot and frames choosing a next project as the first step forward.
- Progress from basics to chatbot: Describe the progression from using the terminal and running short programs to building a locally running AI chatbot with a web interface, while exploring related AI topics.
Why it matters: Recognizing what you have already built can help you choose a realistic next project. - Choosing a next project: Explain what first step the introduction recommends for continuing after the book, regardless of whether the aim is personal use, workplace use, or an AI engineering career.
Why it matters: A concrete project gives your continued learning a direction.
Key terms: terminal, locally running AI chatbot, web interface, path forward, deciding what to build next
18.1 What you have accomplished
This section recaps the book’s layered AI-development skills and the practical independence they give the reader.
- Layered AI development skills: Describe how the book’s skills build from working with local models to building AI applications and understanding more advanced approaches. Give concrete examples from the recap without explaining their implementation.
Why it matters: Seeing how the skills connect helps you recognize the foundation you can build on. - Practical independence with AI: Explain the practical shift the section claims: from using other people’s AI tools to building and evaluating tools for yourself or others. Describe examples of the control this can provide over deployment, sensitive data, and costs.
Why it matters: These skills give you choices about the tools you build and the services you depend on.
Key terms: layered skill set, foundation for AI application development, Ollama, chatbot, RAG architectures, external APIs, local models, proprietary APIs, sensitive data
18.2 Your learning roadmap
Introduces a three-level learning roadmap from the book’s foundational skills toward advanced production work, with a flexible timeline.
18.2.1 Level 1: Foundation (what this book covered)
This section summarizes the local AI foundation the reader has built and explains why it makes further learning easier.
- Transferable local AI foundation: Explain how the working local AI environment and development skills gained in the book prepare the reader to follow new tutorials and extend their projects without relearning the basics.
Why it matters: Your existing setup and skills let you spend more time learning something new and less time getting started.
Key terms: local AI environment, Streamlit chatbot, virtual environments, dependencies
18.2.2 Level 2: Intermediate (the next 3–6 months)
The section maps out intermediate project areas and recommends deepening skills by completing one useful project before exploring the next area.
- One Project at a Time: Explain why the section recommends choosing a project that solves a real problem and completing it before exploring other intermediate topics.
Why it matters: A finished, useful project gives you a stronger base for the next topic than trying to sample everything at once.
Key terms: level-2 learning, RAG, embeddings, retrieval, prompt augmentation
18.2.3 Level 3: Advanced (6–12+ months)
This section sketches advanced directions while emphasizing that open-ended production and community work requires independent judgment and continued learning through iteration.
- Independent advanced learning: Explain how learning changes at the advanced level and what remains the same about the learning process.
Why it matters: It sets realistic expectations for advanced work without making setbacks seem like a break from learning.
Key terms: independent judgment, independent exploration, read, experiment, build, fail, fix, repeat
18.3 Final thoughts
The section closes the book by emphasizing that transferable building skills and continued practice matter even as AI tools change rapidly.
- Enduring AI Building Patterns: Explain why the skills developed in the book remain useful as models, frameworks, and APIs change. Describe the recurring work of connecting a model to an interface, handling responses and state, and debugging errors.
Why it matters: You can adapt to new AI tools by relying on patterns you have already practiced. - Local Models As Fallback: Explain why a local model can be useful when access to a hosted model changes unexpectedly.
Why it matters: Keeping a local option can reduce dependence on a cloud provider’s availability. - Practice Sharing And Curiosity: Describe the section’s three principles for continuing to learn: start building, share what you learn, and stay curious. Explain how practice and teaching help when progress feels difficult or tools change.
Why it matters: You do not need a perfect first attempt to keep growing as an AI builder.
Key terms: fundamentals, underlying patterns, manage state, handle errors, hosted model, local models, fallback, Start building, Share what you learn, Stay curious
18.4 Exercises
Concepts explained in this chapter
- LangChainLangChain is a software framework for building applications with language models by connecting them to data sources and composable components.
- retrieval-augmented generationRetrieval-augmented generation is a technique that retrieves relevant information from an external knowledge source and supplies it to a language model alongside a question or instruction.
- PythonA high-level general-purpose programming language known for readable syntax and a broad software ecosystem.
- local aiLocal AI is artificial intelligence whose models or computations run on a user's own device rather than on remote infrastructure.
- lsA command-line utility that lists the contents of a directory.
- pipelineA sequence of processing stages in which each stage feeds its output to the next.
- embeddingAn embedding is a numerical vector representation of an item, such as text, positioned to capture aspects of its meaning or features.
- TerminalTerminal is the macOS application for accessing a command-line interface.
- virtual environmentA Python virtual environment is an isolated environment with its own installed packages.
- AI agentAn AI agent is a software system that perceives information, reasons about tasks, and takes actions to accomplish goals, including software-development tasks.
- GGUFA file format for storing machine-learning model tensors and metadata for inference, including quantized models.
- GemmaGemma is a family of open language models developed by Google.
- HomebrewHomebrew is a package manager for installing and managing software on macOS and other Unix-like systems.
- LlamaA family of large language models developed by Meta.
- MistralA family of large language models developed by Mistral AI for natural-language processing and generation.
- OllamaOllama is a tool for downloading, running, and serving language models locally.
- StreamlitStreamlit is a Python framework for creating interactive web applications from Python code.
- cd commandA command that changes the current working directory.
- fine-tuningFine-tuning is the process of further training a pretrained model on selected examples to adapt its behavior or capabilities.
- isolationThe separation of a project's packages and runtime from those of other projects or the host system.
- large language modelA machine-learning model trained on large amounts of text to process and generate language.
- macOSApple's desktop operating system for Mac computers.
- parameter countModel size is a measure of a language model's scale, commonly expressed by its number of learned parameters.
- pwdA command that prints the path of the current working directory.
- session contextSession state is application data associated with a user's session and retained across interactions or reruns.
- stack traceA Python diagnostic report showing the sequence of calls leading to an exception and where it occurred.
- system promptA system message supplies instructions that guide an AI model's behavior in a conversation.
- vector databaseA vector database stores vector representations and supports searching for vectors that are similar or close to a query vector.
- Open WebUIOpen WebUI is an open-source web interface for interacting with locally hosted language models and related services.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.