- Beratungsdienste
Chatbot Azure Open AI and ChatGPT MVP
Businesses will unlock the full potential of a Custom Copilot, that uses internally Azure Open AI Service + ChatGPT on Microsoft Teams, combining all the internal tabular data of your company,
Unlock the value of insights generated with AI using Azure Open AI and ChatGPT -4, that can help you make better decisions, identifying patterns and trends of your data to optimize your operations, and having direct communication through one the most used platform Microsoft Teams (other end points can be added on future phases).
With this, customers will have the ability to access your company's indicators and highlights through natural language, gathering information from different sites and files, which may consist of structured, unstructured, and semi-structured data.
OUR PROVEN APPROACH
-Questions can be asked in multiple languages
-It can complement the responses with relevant data when available in the data model (comparisons vs previous year, trends, etc.)
-It can guide you to the source origin of each of the chatbot's answer
-Customer's private data is not exposed, as their data remains within their private clouds and is governed at the database level according to user permissions.
-It is a fully scalable platform in terms of PTUs (Processing Time Units) and can gradually add integrable technologies.
OUR METHODOLOGY
Our 8 - 12 weeks approach consists in 3 main phases
-Pre - requisites (2 weeks): This MVP needs some pre - requisites, such as data source connectivity, a gold layer on the cloud accessible and ready to be consumed, Microsoft Teams and certain validations on the very first 2 weeks.
-Discovery (2 weeks): The next phase, consists of building functional and technical workshops, to understand the data source, indicators, dimensions, aggregations, and to build together the questions that the business users will be doing, to create the context for the chatbot. Also, to do the enablement of the architecture needed, and the governance of the data source that will be consumed. At this phase it is considered just tabular data, but in future phases can be included semi - structured and non-structure data (not included in this scope).
-Development (6-8 weeks): training the models and chatbot, with natural language, doing the questions to the data source and doing all the adjustments needed in the phase takes between 6 and 8 weeks, to do at the end the unit tests and deploy.