Introduction

This document provides practical guidance for IBM consultants, SAP practitioners, and enterprise architects on using IBM Bob, IBM's AI-native engineering assistant, to accelerate and govern SAP development. It covers the full spectrum from initial setup and connectivity through governed code generation, workflow automation, and integration with the broader SAP and enterprise tooling ecosystem.

The focus is on real-world applicability: every section is grounded in hands-on implementation experience across SAP landscapes including S/4HANA, SAP BTP, SAP Integration Suite, and ECC. Readers can find concrete configuration steps, example prompts, architectural guidance, and worked examples that can be adapted directly to client engagements.

For more information and disclaimers about the information and code examples provided in this publication, see the notices section of this document.

Who should read this document

This document is intended for the following users:

• SAP Developers and ABAP Engineers who want to accelerate code generation, refactoring, testing, and documentation using AI-assisted tooling integrated directly into their development environment
• SAP Architects and Solution Architects who need to understand how IBM Bob fits into the broader SAP and enterprise architecture, how governance controls are applied, and how to define standards that the tool enforces across teams
• IBM Consultants and Project Leads who are preparing to introduce IBM Bob on a client engagement and need a comprehensive reference covering both the technical setup and the delivery methodology
• DevOps and Platform Engineers who are responsible for connecting IBM Bob to SAP systems, CI/CD pipelines, and external enterprise tools such as ServiceNow, GitHub, and Tricentis

You should have working knowledge of SAP development concepts (ABAP, BTP, or SAP Integration Suite), familiarity with Visual Studio Code, and a basic understanding of AI coding assistants. No prior experience with IBM Bob is required.

Section overview

Using IBM Bob for SAP introduces IBM Bob and makes the case for its adoption in SAP environments. It covers the market forces driving AI-native engineering, what IBM Bob is and how it differs from generic coding assistants, a comparison with SAP's own Joule 4 Developers tooling, and key development approaches, including Spec Driven Development and Vibe Coding, that shape how the tool is used most effectively.

Connecting IBM Bob to SAP provides the technical foundation for integrating IBM Bob with SAP systems. It covers the available connectivity approaches (ABAP Remote FS Extension, ABAP ADT Extension, and the SAP ADT MCP Server), deployment models, architectural considerations for S/4HANA, ECC, and SAP BTP, and a complete walkthrough for governed AI-assisted development using abapGit, including step-by-step configuration, transport management, and GitHub connectivity.

Working with IBM Bob establishes the operational framework for using IBM Bob in enterprise SAP projects. It covers IBM Bob's built-in modes (Agent, Plan, and Ask) and how to author custom modes for SAP domains, the full permission model, and the Dual-Engine Governance model of Rules and Skills. This section also addresses centralized configuration and distribution for teams, SAP Clean Core and ABAP Cloud standards enforcement, naming conventions, coding standards, and Spec-Kit mode for spec-driven delivery.

Development with IBM Bob provides deep, hands-on coverage of IBM Bob across specific SAP development domains. Topics include ABAP code generation tips and best practices, live code generation via the ADT and ABAP FS extensions, SAP BTP UI5 and CAP development, SAP Cloud Integration (CPI) Groovy script generation and governed architecture, and comprehensive guidance on generating SAP project documentation including functional specifications, technical specifications, and business process procedures.

Integrations and Operation covers IBM Bob's integration with the enterprise tooling ecosystem beyond SAP. It explains how the Model Context Protocol (MCP) enables a single governed interface to multiple systems, then provides detailed integration guides for ServiceNow (change and incident management), GitHub and CI/CD pipelines, Tricentis Tosca (SAP test automation), SAP Signavio and LeanIX (process and enterprise architecture intelligence), and SAP BTP administration. Each integration includes architecture diagrams, authentication configuration, and practical worked examples.

Getting Started provides a hands-on environment setup guide for all three primary SAP development tracks, ABAP, SAP BTP UI5/CAP, and SAP Integration Suite (CPI). It walks through prerequisites, extension installation, MCP server registration, governed rules and skills configuration, and end-to-end verification. The section closes with a Crawl–Walk–Run adoption roadmap that maps the progression from individual developer productivity through enterprise-scale distributed delivery.

Governance and Scale addresses the governance disciplines required to operate IBM Bob responsibly in production SAP engagements. It covers the layered guardrail model (data access, action, model behavior, and audit boundaries), security posture, human-in-the-loop design patterns, and a practical framework for measuring value across immediate, project-level, and strategic dimensions. The section concludes with a 30–60–90 day action plan to move from initial adoption to embedded, scaled delivery.

Appendix A provides the complete source code for the ZDOWNLOAD_ABAP_OBJECT sample ABAP program, which serializes selected ABAP objects with abapGit and downloads the resulting files to a local folder so that IBM Bob can work with them. It is referenced from the Connecting IBM Bob to SAP section.

Authors

Dharma Atluri is an IBM Distinguished Engineer, Global CTO for AI-Led SAP Manage with over 25 years of experience in SAP and with IBM for over 16 years. He is a published author with SAP-Press, Packt, and Archiee delivering both technical and industry-driven books regarding the roles of AI and Cloud based journey for clients. He is also an IBM Senior Inventor with 3 granted patents and overall 18 published patents. He is an active blogger and has experience delivering global enterprise-scale transformation journeys for clients across industries. He has been actively driving the global adoption of IBM Bob for SAP engagements along with ICA Edge and other IBM Tech products.

Tuhin Bhaduri is a seasoned AI Architect and SAP transformation leader with over 22 years of experience delivering enterprise-scale innovation across global organizations. He specializes in helping clients build, adopt, and scale AI capabilities within complex enterprise landscapes, particularly across the SAP ecosystem. Currently, Tuhin supports multiple enterprise accounts in designing and implementing AI-driven business solutions that accelerate operational efficiency, intelligent automation, and digital transformation. His expertise spans Agentic AI, watsonx Orchestrate, IBM Consulting Advantage, and IBM Bob, enabling organizations to create intelligent, enterprise-ready AI solutions. Tuhin brings deep functional and technical expertise across SAP technologies, including SAP BTP (BAIP), SAP S/4HANA, SAP Signavio, and LeanIX. His extensive background in enterprise SAP development and transformation allows him to bridge business strategy with modern AI and automation capabilities, helping organizations modernize processes and unlock measurable business value. With a strong combination of architectural vision, SAP domain expertise, and AI innovation experience, Tuhin is recognized for driving impactful enterprise transformation initiatives across industries.

Shubhangi Joshi is a seasoned SAP Architect with over 20 years of experience delivering enterprise-scale SAP transformation programs. A published SAP Press author of "ABAP on Cloud," she is an expert in SAP S/4HANA, SAP Business Technology Platform (BTP), SAP Business AI, Clean Core, and cloud-native development. Passionate about innovation, Shubhangi helps organizations accelerate their digital journeys through scalable and future-ready solutions. Her core specializations include enterprise architecture, SAP extensibility, Generative AI, Agentic AI, and modernization strategies. Known for her high energy and collaborative leadership, she is a frequent contributor to technical communities, knowledge-sharing initiatives, and enablement programs across the IBM and SAP ecosystems. Her professional mission is to bridge business challenges with emerging technologies, driving sustainable value and team growth through a customer-first mindset.

Phillip Gerrard is a Project Leader for the International Technical Support Organization working out of Beaverton, Oregon. As part of IBM® for over 20 years, he has authored and contributed to hundreds of technical documents published to IBM.com and worked directly with IBM's largest customers to implement storage solutions and resolve critical situations. As a team lead and Subject Matter Expert for the IBM Spectrum Protect support team, he is experienced in leading and growing international teams of talented IBMers, developing and implementing team processes, creating and delivering education. Phillip holds a degree in computer science and business administration from Oregon State University.

Meher Charan Vangipurapu is an Architect at IBM with 19 years of enterprise technology experience spanning NLP-based intelligent automation, SAP delivery, and Generative AI. He has served as an Ansible Architect across multiple IBM engagements covering both SAP and non-SAP landscapes, and has applied GenAI tools including IBM Bob, ICA Edge, Ansible Lightspeed, and IBM Consulting Advantage to accelerate delivery using spec-driven and vibe coding approaches. Meher currently focuses on architecting Agentic AI and intelligent automation solutions for SAP ecosystems, helping clients unlock measurable business value through modern AI-assisted development practices.

Mohammad Zubair is a Solution Architect at IBM with 19 years of industry experience, including 16 years at IBM, specializing in enterprise automation, SAP transformation, and Generative AI. He has led the Ansible automation workstream as an Architect for SAP RISE engagements, driving automation across SAP Day 0, Day 1, and Day 2 operations. He has also contributed to NLP- and AI-driven innovations through initiatives such as Ticket Analytics and Catapult Wisdom. Zubair brings strong hands-on expertise across IBM Bob, ICA Edge, Ansible Lightspeed, Ansible Automation Platform, and Agentic AI solution design. He currently leads AI-driven solution architecture and intelligent automation initiatives for SAP clients as part of IBM's AI-Led SAP Manage practice, focusing on modernizing enterprise operations through automation, Generative AI, and agentic solutions.

Feedback

This document reflects practical experience implementing these technologies in production environments. If you encounter issues not covered here, or if you have suggestions for improving the guidance, your feedback helps make this resource more valuable for the community.

Contact IBM Redbooks