In the realm of data orchestration and conversational AI, where the intricacies of integration shape the future of business processes, I embarked on an internship that dived deep into SAP Conversational AI and the integration powers of Jitterbit. This voyage led me from Jitterbit University’s halls to SAP Conversational AI’s interface, culminating in the creation of a sales chatbot with the ability to fetch product information and quote statuses from Salesforce.
Jitterbit Training – Setting the Stage
The first week served as the foundation for this technical odyssey. Jitterbit University courses, including “Introduction to Jitterbit Harmony iPaaS Platform” and “API Creation and Management,” introduced me to the notions regarding an iPaaS Platform and the value it can provide to an organization, a solid understanding of what Jitterbit is, how it works, and what it can do and equipped me with the knowledge to design, deploy, and manage Jitterbit implementations from within Cloud Studio.
Jitterbit is an integration platform that helps businesses connect their systems, applications, and data. It allows users to easily integrate various systems and applications, such as Salesforce, Oracle, SAP or even custom-built applications, by providing a simple, intuitive interface for creating and managing integration workflows. Jitterbit also offers a variety of pre-built connectors, so users don’t have to write any code to integrate their systems. Additionally, Jitterbit provides real-time monitoring and analytics, as well as support for data transformation and mapping.
Moreover, the courses covered several functions and techniques that can be used in Jitterbit Harmony Cloud Studio to create data mappings, an overview of the different ways provided in order to compliment theintegration experience through scripting.
A crucial part of the courses was the API creation and management, which teaches you how to create, run and secure APIs within the Jitterbit iPaaS platform with hands-on assignments and access needed to a REST client like Postman.
Salesforce Data Exploration
With Jitterbit as our orchestra conductor and in between Jitterbit courses, I spent a bit of time exploring and getting acquainted with the Salesforce org used in the internship and also completed a few Trailhead modules. I navigated the intricacies of the Salesforce objects in question, such as Products, Quotes and Quote Line Items and created test data records needed for implementing the queries from Salesforce.
SAP Conversational AI Setup – Laying the Chatbot’s Groundwork
The following week marked the transition into SAP Conversational AI, where I created simple test chatbots primed for interaction. This involved designing conversational flows, defining intents, setting up entities—technical skills essential for crafting a responsive chatbot and creating chatbot skills in which there were implemented triggers, requirements and actions to perform.
The Conversational AI offers a single intuitive interface to train, build, test, connect and monitor chatbots embedded into SAP and third-party solutions, a high-performing natural language processing (NLP) technology and low-code features to ensure faster development.
The Integration
The final and most technically demanding week was dedicated to realizing the integration between SAP Conversational AI, Jitterbit, and Salesforce.
To extract a list of products from Salesforce, we first used Postman—an HTTP client. Leveraging Salesforce’s REST API, we crafted requests to retrieve product data as JSON responses. This laid the groundwork for data retrieval.
Simultaneously, we employed Jitterbit to achieve the same data retrieval task. Using Jitterbit Cloud Studio, we designed integration workflows that fetched product data from Salesforce through its RESTful API. This included configuring connectors, transformations, and orchestrations—crucial elements of Jitterbit’s integration capabilities.
[Above: Get product information (name) workflow in Jitterbit, of which the Read Api Call operation was triggered by the API, using the GET method]
With data in hand, we expanded the chatbot’s capabilities by incorporating a question that allowed users to inquire about product details based on product names. This involved tweaking chatbot’s intents, entities, and responses to accommodate the new functionality.
Within Jitterbit, we created a custom API that served as the bridge between the chatbot and Salesforce. This API defined endpoints and actions necessary for querying Salesforce’s REST API and returning relevant product information.
An operation, a scripted sequence of actions, was crafted within Jitterbit to execute when the chatbot invoked the API. This operation included making HTTP requests to Salesforce, transforming the retrieved data, and delivering responses back to the chatbot.
As data flowed through the integration, meticulous debugging was carried out using Postman. We scrutinized API requests and responses, ensuring that the entire interaction between the chatbot, Jitterbit, and Salesforce was seamless and error-free.
Instead of Conclusions
In the end, we had meticulously created two workflows within a Jitterbit Harmony project, each tailored to a distinct use case. We then exposed Jitterbit’s operations as a managed API, ready to be invoked by the SAP Conversational AI chatbot. The chatbot seamlessly interacted with Salesforce through Jitterbit’s orchestrated integration, responding to users’ queries and fetching real-time data.
This internship was a technical voyage that transcended traditional Salesforce administration. It uncovered the complexities of AI-driven chatbots, the intricacies of Jitterbit integration, and the fusion of these technologies in a dynamic business landscape. The end result was a testament to the power of technical proficiency and a deep understanding of data orchestration in today’s digital commerce arena.
Last but not least, I want to take this chance to thank the incredible team that stood by my side throughout this internship, offering their guidance and expertise. I’m talking about the coordinators, namely Viorel, Sorin and Cosmin, which are experienced seniors in the fields of Jitterbit, Salesforce, and SAP, by means of which they made this journey not only possible but immensely enriching, and also about my internship colleague, Razvan, with whom I collaborated tirelessly with dedication, teamwork, and shared enthusiasm which were instrumental in turning theoretical concepts into tangible reality.