Data Analytics / Amazon QuickSight / Machine Learning
Establish a centralized and standardized process for the development of more complex dashboards and reports, aiming to unify the tools and platforms used and provide real-time access to the results for consumption.
We implement new processes, perform migration and centralization of dashboards and reports, and build new panels using Machine Learning algorithms integrated into Amazon QuickSight.
These actions have significantly contributed to improving the quality and speed of the information made available for business decisions.
Machine Learning / Cloud / AWS / Amazon SageMaker / Amazon Textract / Amazon Rekognition / Amazon Comprehend
Perform enhancements in the process of manual data analysis and availability in the system. Optimize the time for evaluation and validation of documents more efficiently.
Developed a highly efficient model capable of analyzing, validating, classifying, enhancing, cropping, and automatically extracting text from images, resulting in a significant reduction in the time and effort required to perform these analyses, as well as improving the quality of the results.
Data Analytics / Cloud / AWS / Data Lake / AWS Glue / Amazon RedShift / Amazon Comprehend
Reducing costs in the cloud, creating a suitable and scalable data environment to meet the company's needs, implementing more effective data governance, and centralizing partner data.
Deployment of a serverless Data Lake on the AWS cloud platform, with simplified integrations and centralization of data from various sources and partners through APIs.
Machine Learning / Cloud / AWS / Amazon SageMaker / AWS Glue / Step Functions / Amazon Neptune
Development of a course recommendation engine, through the analysis of data from the institution's student database.
The data necessary for the development of this project was provided by the educational institution, with the support of public data.
Development of a course recommendation engine, through data analysis. Tables of cities, municipalities, IBGE, areas of knowledge, correspondence between areas and courses, and year of completion were considered. To facilitate the creation and execution of graph-based applications, we used Neptune.
Machine Learning / Cloud / AWS / Amazon SageMaker / AWS Data Migration Service (DMS) / AWS Glue / AWS Athena / Amazon QuickSight
Identify new patterns of consumption variation, categorize consumers, automate risk analysis, and expand the product portfolio.
We developed Machine Learning models to identify new consumption variation profiles, categorize consumers, automate risk analysis, and expand the product portfolio. We used public data from CCEE (Electric Energy Trading Chamber) to classify consumers' consumption variation profiles based on their CNPJ (Brazilian corporate tax identification). Then, we applied metrics provided by the client to establish rewards, discounts, or carry out targeted marketing actions for these customers.
Machine Learning / Cloud / AWS / Amazon SageMaker / AWS Glue / Step Functions / Amazon Neptune
The objective of this project was to detect the likelihood of student dropout through the analysis of various available data.
We used Machine Learning model training techniques to assign each student a probability of dropout score based on their profile. The data used was provided by the educational institution and included financial information, attendance records (entry and exit times), and the student's academic history at the institution.
English 
Português do Brasil 
Español