onedevatlasBeta
Browse the atlas
AllBTPCAPUI5ABAP
SampleSampletypeA worked example you read or clone. Evidence: it demonstrates, it never provides.See every one →What this type is →by SAPby SAPoriginPublished by SAP itself. A visible facet, never an input to a reading — a dormant official plugin reads exactly like a dormant independent one.See every one →Reference

SAP-samples/generative-ai-codejam

A CodeJam that deploys and queries a large language model on SAP AI Core's Generative AI Hub, then builds a retrieval-augmented chatbot on top of it.

Exercises deploy an LLM on SAP AI Core and query it through SAP AI Launchpad, the orchestration service and the Python SDK, then create embeddings for document chunks, store them, and use them to implement the retrieval augmented generation case the CodeJam is built around. Later exercises work through the orchestration service and its grounding, a chatbot with memory, and a chatbot the participant writes. The last item, numbered 12, is marked optional and links out to a community blog post on AI agents instead of being an exercise in the repository.

GitHub60 ★

Facts

Last pushLicenceApache-2.0

Where it sits

Lives in
BTP/BAIP
Demonstrates
LLM integrationBTPLLM integrationcapabilityCalling a language model from application code with the tenant, destination and credential handling already solved instead using an SDK call with a hand-rolled key.Retrieval-augmented generationBTPRetrieval-augmented generationcapabilityGrounding a model's answer in data retrieved at question time: text split into chunks, each chunk embedded as a vector, and the chunks nearest to the question placed into the prompt. Semantic search is the same retrieval with a person reading the matches, and fine-tuning changes the model's weights, not its input.
Requires
SAP AI CoreSAP AI CoreartifactThe service Generative AI Hub runs on and the LLM is deployed to; the first exercise connects it.SAP AI LaunchpadSAP AI LaunchpadartifactOne of the three ways the exercises query the deployed model, and where the prompt editor exercise is done.

Who is behind it

SAP SamplesSAP SamplesOrganisationMaintains this
In the atlas: maintains 222@SAP-samplesmaintainerAjinkya PatilPersonContributes to this
In the atlas: contributes to 112@ajinkyapatil8190Antonio MaradiagaPersonContributes to this
Developer Advocate
In the atlas: contributes to 10@ajmaradiagaKevin RiedelsheimerPersonContributes to this
In the atlas: contributes to 11@KevinRiedelsheimerShrinivasan NeelamegamPersonContributes to this
In the atlas: contributes to 4@neelamegamsNora von ThenenPersonContributes to this
In the atlas: contributes to 3@noravthVitaliy RudnytskiyPersonContributes to this
In the atlas: contributes to 6, wrote 1@Sygyzmundovych6 contributors

What it declares

Every artifact

Type
Sample
Origin
By SAP
Use as
A reference

As a sample

Form
Workshop or CodeJam material

Connections4

README

quoted from SAP-samples/generative-ai-codejam under Apache-2.0, read
SAP-samples/generative-ai-codejam/README.mdOn GitHub

CodeJam - Getting started with Generative AI Hub on SAP AI Core

REUSE status

Description

This repository contains the material for the "Getting started with the Generative AI Hub on SAP AI Core" CodeJam, brought to you by the Developer Advocates at SAP.

Overview

In this CodeJam you will learn how to use Generative AI Hub on SAP AI Core to implement a retrieval augmented generation (RAG) use case to improve the responses of large language models (LLMs) and reduce hallucinations. You will learn how to deploy an LLM on SAP AI Core and query it via SAP AI Launchpad, the orchestration service and the Python SDK. Furthermore, you will learn about the most important genAI concepts and create and use text chunks and embeddings to improve your RAG response.

Requirements

The requirements necessary to complete the exercises in this repository, including hardware and software specifications, are outlined in the prerequisites file.

Exercises

You can find all exercises in the exercises folder. We will work through the exercises in the order shown here. From a session flow perspective, we are taking the "coordinated" approach:

The instructor will start you on the first exercise, and that's the only one you should do. You should only proceed to the next exercise once the instructor tells you to.

  1. Connect SAP AI Core and SAP AI Launchpad
  2. Explore the Prompt Editor in SAP AI Launchpad
  3. Setup your Python environment
  4. Prompt an LLM
  5. Create embeddings for your document chunks
  6. Store embeddings
  7. Implement the RAG use case
  8. Orchestration service S3 grounding (UI)
  9. Orchestration service
  10. Orchestration service - grounding
  11. Chatbot with Memory
  12. Your Chatbot
  13. OPTIONAL: AI Agents

Feedback

If you can spare a couple of minutes at the end of the session, please provide feedback to help us improve next time.

Use this Give feedback link to create a special "feedback" issue, and follow the instructions in there.

Thank you!

Other CodeJams

CodeJam repositories
CodeJam Community

Acknowledgements

The exercise content in this repository is based on different sample repositories created by the SAP AI Core team.

The exercises related to the prerequisites and setting up BAS were copied and adjusted from my colleague Vitaliy's CodeJam - Getting Started with Machine Learning using SAP HANA and Python.

How to obtain support

Create an issue in this repository if you find a bug or have questions about the content.

For additional support, ask a question in SAP Community.

License

Copyright (c) 2024 SAP SE or an SAP affiliate company. All rights reserved. This project is licensed under the Apache Software License, version 2.0 except as noted otherwise in the LICENSE file.

Apache-2.0, read

Details

Sources3
Added by handdiscovered herechecked
github.com/SAP-samples/generative-ai-codejamtree, pushes, contributors, CIchecked
README.mdquoted, and what the entry declareschecked

Sources are graph entities: the same page may back several entries, and an unreachable source flags every row that rests on it.

Compatibility and provenance
discovered viaAdded by hand
first seen in the atlas
ecosystem and domainbtp, ai-and-agents
Contributors6
Ajinkya PatilAntonio MaradiagaKevin RiedelsheimerShrinivasan NeelamegamNora von ThenenVitaliy Rudnytskiy

From the repository’s contributor list, .

History in the atlas1
  1. First seen.
For your agent
Ref
artifact:sap-samples-generative-ai-codejam
JSON
/artifacts/sap-samples-generative-ai-codejam.json
MCP
Connecting an agent