The BOSS initiative aims to support underrepresented groups in the tech industry who seek mentorship and guidance to learn how to contribute to open-source projects. We are proud to have been awarded first place in the Gnome Community Challenge.

Our program offers a safe and supportive environment along with a structured methodology to help participants grasp the fundamental concepts of collaborating on open-source projects. By the program’s end, participants will have acquired the necessary skills to continue their contributions to other projects confidently. Our mentoring approach not only addresses technical knowledge gaps but also fosters confidence and engagement among participants. In addition to technical training, the BOSS program addresses inclusion challenges such as intersectionality and impostor syndrome.

For more information, you can access the BOSS initiative manual here.

Awarded by Gnome Community Challenge, Mentoring Program, BOSS, Open Source, Underrepresented groups

The Brazil Participatory Platform is the largest social participation platform of the federal government of Brazil, with over 1.5 million users in 3 months. It is built to allow the population to contribute to the creation and improvement of public policies. The platform was developed using open-source software with the support of Dataprev, the Decidim-Brasil community, the Ministry of Management and Innovation in Public Services (MGI), and LAPPIS.

The OSS Brasil Participativo platform customizes participatory processes from the Decidim framework to suit the Brazilian context. Presently, the Decidim Brasil community boasts over 200 participants, while the OSS Brasil Participativo has garnered the involvement of over 100 direct and indirect contributors in its first year alone. Brazil stands out as one of the primary adopters of this tool on a national level, given its compelling features, adaptability, cost-effectiveness, and steadfast commitment to transparent participatory processes.

Largest Initiative in Brazil in Social Participation, Decidim, rails, data engineering

The DermaCam project aims to implement an innovative strategy through a skin lesion screening optimization system, integrating records and data into an analytical platform following the Data Fabric model for the early diagnosis of skin cancer in Primary Health Care (PHC). This approach is combined with the application of a clinical screening questionnaire, the capture and recording of suspicious images using a dermatoscope, and the integration of the historical series of registered medical visits with ICD codes related to the disease.

Foundation for the Support of Research of the Federal District (FAP-DF), University of Brasilia (UnB), Early screening for skin cancer, Continuing Education, Health Planning, Screening, Skin Cancer, Data Fabric, Machine Learning

The project, developed in collaboration between the Laboratory of Free Software Competence (Lab Livre) at the University of Brasília (UnB) and the Institute for Applied Economic Research (IPEA), focuses on improving the management and integration of structural government data systems in Brazil. These systems, essential for public administration, store critical information on budgeting, personnel, and organizational structures. However, data redundancy, inconsistencies, and interoperability barriers have hindered evidence-based decision-making and transparency. To address these issues, the project developed a technological solution that integrates and qualifies data, ensuring better access, reliability, and usability for public managers and society. Using open-source technologies and automation, the initiative enables the efficient extraction, transformation, and visualization of government data. The project streamlines administrative processes and strengthens public accountability by structuring workflows and enhancing data interoperability. Adopting open technologies reduces dependency on proprietary software, fostering innovation and adaptability across different government entities. With a replicable and scalable approach, the project enhances data-driven governance and promotes a culture of transparency and informed decision-making within the Brazilian public sector.

Institute for Applied Economic Research (IPEA), University of Brasilia (UnB), Data Management, Open Source

The project aims to establish a monitoring system for both the supply and consumption of electricity throughout the University of Brasília (UnB). Additionally, a recommendation system will be implemented to suggest the most suitable contracts for major consumers. This initiative forms part of a broader investment by the University of Brasilia aimed at positioning it as the fourth most sustainable university in Brazil.

Measurement equipment has been strategically installed at various locations within the university to gather diverse indicators pertaining to the quality and availability of electricity. Subsequently, this data will be transmitted across the network to a centralized system responsible for storage and accessibility, serving purposes such as research or ongoing monitoring.

The LAPPIS team leads the the development this open source project of a distributed system tasked with data collection from the meters, its centralization, and delivery to an application designed for user-friendly visualization of information. This visualization will encompass the use of reports, graphs, and comparative analyses, thereby enhancing comprehension and facilitating data interpretation.

University of Brasilia the 4th most sustainable university in Brazil, IoT, Distributed Systems, Energy Monitoring, DevOps, DataViz

The “Free Software Ecosystems” project is a partnership between LAPPIS and the Ministry of Culture (MinC), initiated in October 2017 with a planned duration of 24 months. The partnership aims to bring innovation to the open-source software systems developed by the Ministry and to conduct applied research in continuous delivery, DevOps, and machine learning techniques.

One of the research endeavors related to this project led to the publication of “A survey of DevOps concepts and challenges”, an article with over 450 citations.

The EJ (Empurrando Juntas) is an opinion collection platform that utilizes artificial intelligence techniques to group participants based on their opinions. The grouping is done through personas, which allows for the discovery of new nuances and groups that would not be identifiable otherwise. With the results of a collection, EJ enables understanding of what the target audience thinks about a particular topic, thus strengthening its communication strategy.

In this project, we experiment with several machine learning algorithms to detect outliers and anomalies in large Customer Dataset. We deploy Data, Feature Engineering, and MLOps techniques and practices.

We adopted a commonly used machine learning workflow depicted in various forms across industry and research. It has commonalities with prior workflows defined in data science and data mining, such as TDSP, KDD, and CRISP-DM. They have in common the data-centered essence of the process and the multiple feedback loops among the different stages. We followed some DevOps and MLOps best practices and automation to speed up the experimentation cycle and scale the process to other databases and a larger team.

Machine Learning, DevOps, Data science

Data Visualization Project for Promoting the Cultural Landscape Supported by the Incentive Law. The aim is to generate dashboards illustrating the cultural panorama based on aggregated data from projects submitted to SALIC (the Brazilian Ministry of Culture’s online system), including aspects such as the geographical distribution of cultural projects, free access availability, health index, among others.

Dataviz, Continuous Delivery, DevOps, Cultural Landscape

Project to generate recommendations in the evaluation and submission of cultural projects in SALIC. The objective is to apply machine learning techniques to extract patterns from the historical data of previously submitted cultural projects (approximately 30,000 projects), and to use these patterns to recommend anomalies, areas for improvement in cultural projects, and to generate metrics of complexity and quality for such projects. We work on a microservices architecture to bring innovation to legacy systems (such as SALIC).

Aprendizagem de máquina em projetos culturais, SALIC, Arquitetura Microsserviço, DevOps, Entrega Contínua, SALIC-API

Project of a FAQ chatbot to answer questions about the Incentive Law and assist applicants in filling out a cultural proposal in SALIC. We contribute to various open-source software projects in this initiative, including Rocket.chat and Rasa. We employ deep learning techniques for both natural language processing and dialogue flow learning.

Local OSS community with a chatbot configured and trained for Portuguese language, boasting over 300 members on the Telegram channel, 200 forks, and 200 stars on GitHub.

Chatbot, FAQ, SALIC, RASA NLU,RASACORE, Rocket.chat, Deep Learning, DevOps, Entrega Contínua, ElasticSearch.