Enterprise Natural Language Processing with LLM: Advanced Architectures for Production-Ready AI in Python (The Enterprise Data Science Mastery Series) - Brossura

Halversen, Bjorn

Brossura

ISBN 13: 9798170565047

Casa editrice: Independently published, 2026

Riassunto

Building an LLM prototype is easy. Engineering one that survives production is not.

Enterprise Natural Language Processing with LLM is a practical architectural guide for software architects, AI engineers, and developers building production-ready NLP and LLM applications in Python.
Move beyond simple prompts and demos to understand the systems, infrastructure, security, and engineering practices required to operate sophisticated generative AI at enterprise scale.
Inside, you'll learn how to:

• Architect LLM-powered applications, including agentic and conversational systems
• Design advanced RAG architectures with hybrid search, re-ranking, query transformation, and graph-based context
• Manage prompts, context windows, token limits, versioning, and complex workflows
• Build and orchestrate multi-agent systems in Python with routing, human-in-the-loop validation, and execution tracing
• Control LLM costs through semantic caching, model routing, batching, and API optimization
• Fine-tune and self-host open-source models using LoRA, QLoRA, vLLM, TGI, and TensorRT-LLM
• Defend AI applications against prompt injection, jailbreaks, data leakage, and other security risks
• Establish LLMOps pipelines for evaluation, A/B testing, model management, and continuous deployment
• Design model-agnostic architectures capable of adapting to evolving AI technologies

Whether you're designing an enterprise chatbot, RAG platform, autonomous AI workflow, or broader generative AI infrastructure, this book focuses on the architectural decisions that matter when prototypes become production systems.

Build LLM applications that are scalable, secure, observable, and ready for the demands of real-world enterprise AI.

Dettagli dell'edizione

• Editore: Independently published
• Data di pubblicazione: 2026
• Lingua: Inglese
• ISBN 13: 9798170565047
• Rilegatura: Paperback
• Numero di pagine: 144
• Contatto del produttore: non disponibile
• Persona responsabile: non disponibile