Enterprise AI × SAP
Learn enterprise AI the way SAP teams actually use it.
Enterprise AI for SAP, explained twice: the essentials first, then the deep dive. Every topic has a foundational layer and a deep layer.
New: practice configuration in Orchestrate Lab, from an empty system to a closed month at a fictional pump maker.
172 topics live · 36 new or updated in the last 7 days
Curriculum
Nineteen learning units and thirteen portfolio projects (8 live). Each unit pairs an AI concept with how SAP implements it, and ends with a skill a forward deployed engineer uses with clients.
-
1
Foundations
What an SAP forward deployed engineer does, where AI sits in SAP, and the engineering basics every later unit builds on.
-
1B
SAP end to end
How an SAP enterprise is wired: org structure, shared master data, document flow and every major process across modules, plus the solution architect's design decisions (landscape, fit-to-standard, roles, data migration, analytics and non-functional requirements), so you can lead design and integration on a real program.
-
2
Machine learning fundamentals
How machines learn from examples, the metrics that decide whether a model is useful, and a first prediction on SAP-shaped data.
-
3
Neural networks and embeddings
Neural networks from scratch, how embeddings capture meaning, and semantic search over SAP master data.
-
4
Attention, transformers and LLMs
The machinery inside large language models, built small enough to understand end to end.
-
5
LLM engineering
Choosing and calling models, prompt and context engineering, structured outputs, and SAP's Generative AI Hub.
-
5B
Integration architecture
How SAP and non-SAP systems exchange data, the integration platform and clean core, and how to run the integration workstream of a program, with AI agents as one more consumer.
-
6
AI applications
Turning a model call into a service, extending SAP side by side, and deploying on BTP.
-
7
Retrieval-augmented generation
Grounding answers in your documents and SAP data, with retrieval that respects authorizations.
-
8
Evaluation
Measuring whether an AI system is right, proving it to the business, and catching regressions.
-
9
Agents and MCP
Agents from first principles, tool design, the Model Context Protocol, and Joule agents.
-
10
Production and LLMOps
Observability, cost, latency and operations for AI systems, and the business case that funds them.
-
11
AI security and governance
How AI systems get attacked, how to defend them, how SAP authorizations apply to agents, and how companies govern AI use.
-
12
Models and advanced AI
When to fine-tune, local and small models, document AI, tabular AI, and SAP's embedded AI.
-
13
AI-assisted SAP engineering
Coding agents, ABAP Cloud and RAP, and agents that call SAP business logic safely.
-
14
The FDE operating system and capstone
Discovery, opportunity assessment, executive communication, adoption, and the capstone engagement.
-
16
AI across SAP's cloud suite
Where SAP ships its named assistants and agents outside S/4HANA: SuccessFactors, Ariba, Concur, CX, IBP, Signavio, LeanIX and Build. What each does, its status, and what the consultant changes.
-
17
SAP AI next to other platforms
Most customers run SAP beside Microsoft, AWS, Google, Salesforce or ServiceNow agents. How those platforms reach SAP, where an agent should live, and what the API policy allows.
-
18
Industry playbooks
How AI use cases, constraints and SAP's own agents differ by industry. Each playbook starts from the processes a consultant already knows.
-
P
Portfolio projects
End-to-end projects that take SAP data all the way to a deployed, evaluated solution with a business KPI. Each one is a portfolio piece.
Latest topics
- Multi-agent supervision and handoffsNov 18, 2026
- Set up for Unit 6: app tooling: Node.js, CAP and Cloud FoundryNov 17, 2026
- Building an AI APINov 17, 2026
- CAP and side-by-side extensions for AINov 17, 2026
- Agent memory and long-running agentsNov 17, 2026