Skanda Vivek, an assistant research scholar at the Center for Governance of Technology and Systems (GoTech), has released a new book with O’Reilly Media titled Retrieval-Augmented Generation in Production with Haystack.
Vivek is a senior data scientist on the Risk Intelligence team at OnSolve, where he develops AI-based algorithms for detecting critical emergencies using big data. His background includes academic appointments at Georgia Gwinnett College and Georgia Tech, and he holds a Ph.D. in physics from Emory University.
The book offers a practical guide to implementing retrieval-augmented generation (RAG) systems using Haystack, an open-source framework that simplifies the development of scalable, API-driven large language model (LLM) backends. It introduces software engineers to advanced capabilities such as hybrid RAG, semantic search, SQL integration, and the development of ChatGPT-style tools on custom data.
Retrieval-Augmented Generation in Production with Haystack provides machine learning and backend engineers with a hands-on resource for deploying LLM-powered applications in production environments. The publication aligns with GoTech’s mission to support cutting-edge research at the intersection of emerging technologies and public policy.
