Generative AI with Python: The Developer’s Guide to Pretrained LLMs, Vector Databases, Retrieval-Augmented Generation, and Agentic Systems

Building useful generative AI applications involves more than sending a prompt to a language model. Generative AI with Python takes a developer-focused path through pretrained models, prompt engineering, vector databases, retrieval-augmented generation, and agentic systems, with Python coding examples connecting the ideas to implementation.

From AI foundations to working models

Bert Gollnick opens with an introduction to generative AI and its relationship to natural language processing and large language models. The book then turns to pretrained models and practical tasks such as summarization, classification, question answering, and named entity recognition. This progression gives readers a foundation before they begin assembling more involved applications.

Prompting, chains, and model choices

Dedicated chapters explore prompt engineering and working with language models through Python. Along the way, the material addresses practical considerations such as model hosting and the differences among open-source, open-weight, and proprietary models. These topics help frame implementation decisions alongside the code itself.

Make documents searchable with vector databases

The vector database material follows the process from loading and chunking text to creating embeddings, storing data, and retrieving relevant results. Examples include Chroma and Pinecone, giving readers a way to examine different approaches to organizing and querying information.

Connect retrieval to generation

Retrieval-augmented generation brings a model together with an external knowledge source. The book covers a basic RAG workflow as well as more advanced retrieval stages and evaluation challenges. This makes the chapter useful for developers interested in grounding model responses in document collections.

Explore agentic systems

The final stretch introduces AI agents and frameworks including LangGraph and CrewAI. Coding sections and exercises appear throughout the contents, making the book a hands-on technical read for Python developers exploring how generative AI components fit together in applications.

With its progression from pretrained models to RAG and agents, this guide offers a clear structure for readers who want to understand both the building blocks and the application patterns behind generative AI with Python.