You’ve Arrived at Bot Games Season 4 - AI Edition!
Welcome to the first challenge of Bot Games Season 4 - AI Edition! Who doesn’t love some friendly competition? We're challenging you to showcase not only your automation building abilities, but also your proficiency in crafting well-designed prompts of generative AI models that elicit the desired responses in the appropriate format. The spotlight is on the potential of Generative AI, and we're excited to witness how you can harness that power to develop integrated automation solutions. So get ready to flex your creative muscles and show us what you've got.
New to Bot Games? Bot Games is a global competition that challenges developers to build automations that can solve realistic business challenges in free-to-use, scored, and timed environments. Head to our Bot Games HQ here for more information.
Challenge #1 Introduction
It's time to get your creative juices flowing and reimagine the way you've previously thought about automation! Our first challenge is a forward-looking use case based in the healthcare industry that demonstrates the power of integrating Generative AI within the Automation Success Platform using strictly fake data. If you are considering automating use cases that contain actual PHI or PII with Generative AI, please consult your legal team before proceeding.
For more details and tips to get started, visit the Challenge #1 Use Case page.
Challenge #1 Objective
A healthcare company, Medikorps, provides treatment for a wide variety of patient conditions and wants to standardize its report-generation process. Once a patient visits with a doctor, the doctor analyzes the patient and shares his/her notes about the patient's condition along with prescribed medications if needed. The doctor then sends an email to the medical assistant to generate a medical report to be shared with the patient. As the healthcare facility receives a large number of patients every day, generating reports has become a bottleneck. Therefore, the facility intends to implement an automation solution to streamline the process.
Your automation should launch the challenge page, then launch the mailbox from the challenge page, as both are interlinked. Then, send the information in the doctor's notes to a Generative AI provider of your choosing (ChatGPT for example) to extract the required details in a format that can be easily interpreted by the automation.
The mailbox will have 10 emails in total, of which only a few will be unread. Data needs to be extracted only from the unread emails to be entered into the report generation application. Note that the unread emails will change every time you open the mailbox.
You should select the 'Medication Prescribed' checkbox only if the doctor has prescribed any medication.
Challenge #1 Tips
- ChatGPT can respond back in different formats. Try phrasing prompts in different ways to see how the response varies. Be specific about the values for which you are looking and the format in which you want the response.
- If you plan to use JSON format, this video on how to parse JSON using the JSON package is available.
Challenge #1 Solution Tutorial
An effective way to incorporate Generative AI into intelligent automation is to extract and summarize data into a standardized format. Generative AI can be utilized to extract specific information from long-form unstructured text, which your automations can then leverage to automate complex end-to-end workflows. This technique can be highly beneficial in various domains, such as extracting information from legal documents, research papers, and other documents with unstructured data. By implementing Generative AI with automation, you can improve the accuracy and efficiency of your workflows, while reducing manual labor and minimizing errors. With the vast amount of unstructured data that exists today, the potential for this approach is virtually limitless.
- Using the cutting-edge OpenAI Generative AI Package brings the power of OpenAI's advanced generative AI technology directly to the Automation Success Platform. It functions with GPT-4, ChatGPT-3.5 Turbo, and GPT-3 models with both OpenAI and Microsoft Azure service providers. Check out this link to learn more about this package and its capabilities.
Also, one of the enjoyable aspects of building automations is discovering the various methods of solving the same problem. You can try using different packages or actions to complete the challenge and observe how those changes affect your automation’s processing time, accuracy, and stability.
View the full solution tutorial video here and get more details on the various packages and approaches that can be utilized to solve this challenge.