Carlos – JavaScript, React, Node.js

Muse is an AI-powered online platform developed by FormationBio to support pharmaceutical drug development studies. Leveraging advanced AI capabilities (OpenAI), Muse analyzes the data provided about new drug development projects and generates comprehensive overviews tailored to the user's needs.

Key Features:

• Data Analysis and Summarization: Processes and refines complex study data into clear, actionable insights.
• Target Recommendations: Suggests the most relevant groups, patient targets, and platforms for advertising the drug.
• Strategic Options: Provides tailored suggestions to maximize outreach, helping pharmaceutical companies identify the best possible patients for their studies.

Who It’s For: Muse is designed for pharmaceutical companies, researchers, and drug development teams looking to optimize the early stages of study promotion and patient recruitment.

• Updated the front-end UI to align with new Figma designs provided by the design team.
• Implemented social media ad previews (Facebook, Reddit, Google) for users to visualize live content.
• Built a parser function to dynamically attach citation previews by analyzing React components.
• Enhanced the component library by:
  • Adding new components and fixing existing ones.
  • Refactoring all tests to replace Enzyme with React Testing Library.

This is a web application that integrates blockchain technology to create a unique experience centered around a cryptocurrency. It allows users to earn the crypto by watching in-app ads or playing games. Users can also gain tickets to enter weekly, provably fair jackpots.

Their referral program rewards users for bringing new members into the community. A portion of the revenue generated from ads and in-app purchases is directed to support environmental causes. Additionally, EveryWorld is part of a broader ecosystem that includes two other projects:

• A platform where users can play various games using the crypto.
• A casino where users can gamble with this or other cryptocurrencies.

• Implemented a dashboard for users to view crypto metrics
• Led the development of an MVP mobile app using React Native;
• Updated the UI to enhance the user experience and animations;
• Rebuilt the UI for onboarding and the logic for in-app notifications;
• Refactored core features, improving the performance of the video player and scratch card game;
• Set up Storybook for the new design system shared between the apps;
• Wrote unit tests.

A smart platform with CRM, prospecting, and post-sale management. Bringing automation throughout the process to ensure that the client finds leads with the best conversion chances. The client can rely on proposal email management, WhatsApp outreach, and even AI assistance to help with decision-making during negotiations.

• Implemented a new UI with optimized UX to facilitate the user journey during the prospecting stage;
• Optimized and refactored bottleneck points, reducing delays in critical prospecting flows;
• Improved the coverage of unit tests and end-to-end tests to ensure more stable releases;
• Debugged and solved issues with the automated credit distribution feature;

A company from Spain that develops and distributes platforms that enable remote learning for schools. As a consultant, Carlos supported the adaptation of their platform for Brazilian schools, implementing features necessary to meet their specific needs.

Carlos also contributed to another project for the company — a platform designed to assist students in their studies using AI. The platform featured a chatbot powered by Langchain and ChatGPT, allowing students to ask questions about their lessons based on PDFs provided by their teachers. Additionally, teachers could use the platform to generate quizzes and automatically assess student responses.

• Completed the platform’s dynamic translation using i18n for localization to Brazilian schools;
• Implemented specific features to cater to the Brazilian education system;
• Debugged and resolved issues raised by the quality team during adaptation;
• Developed a backend API to parse lesson PDFs provided by teachers, categorizing the content to deliver accurate responses through the AI chatbot;
• Created APIs to generate questions based on lesson content and integrated them into the platform;
• Maintained and optimized the chatbot API, improving the user experience for students interacting with the AI.

The product is a web application created to improve store mobile app rating scores through more humanized customer service.

The tool scrapes the client's app reviews, gets the most recent, and sends it to attendants who read, catalog, and answer the reviews in a way that makes the users feel heard. It has shown excellent results and helped the customer service of most of the top 500 apps in Brazil answer thousands of reviews daily.

• Refactored the back end to use class composition;
• Refactored the data layer of the front end to remove Redux where it wasn't needed;
• Implemented React Query to improve speed and reduce the number of unnecessary requests;
• Helped with breaking the monolith API into microservices;
• worked with the UI/UX team on the research on building their component library focused on accessibility and performance;
• Implemented a real-time service to track and block user actions based on other users' actions (to avoid concurrency on some critical pages);
• Implemented a real-time service that allows users to share the tickets to events they're attending and collaborate.

It's a questions manager where the users can create questions, assign them to reviewers, keep track of changes, create tests for their courses, and send them to the students.

The goal of this project was to centralize the information and retire the spreadsheets used. Some essential features included role-based access to the information and automatic generation of the test's Word document with the test data. And the most important of them all, a git-like changes history to track who changed what and when.

• Worked in the back-end team to rewrite the GraphQL module to make it more scalable and maintainable;
• Fixed security and performance issues such as the n+1 query problem and the infinity of nested queries;
• Worked closely with the UI/UX team on implementing the new UI of the platform to make it more modern and provide a better user experience.

A corporate education provider. To deliver good quality content for their clients, they needed to hire the best contractors in different areas, such as translators, writers, designers, etc. They started this project to make the process easier.

It's a web application where the candidates can sign in and fill out their information. Based on some key fields, the system will search for the candidate data on other company applications and aggregate it. HR can find the most compatible profiles when they need a contractor with a particular skill set.

• Implemented a data aggregation API;
• Developed a fully functional and flexible filter module that allows the HR to make more complex queries by creating dynamic filters based on the data in the database;
• Improved the back end by refactoring some slow queries that used to consume a lot of resources to compute;
• Worked on improving the security and usability of the external form that the contractors could use to update their data (one of the main responsibilities was refactoring it to a mobile-first approach in the UI);
• Communicated with the stakeholders (as the lead developer) to ensure that the features we were developing were the right fit for their needs.

A project for a local cycling association that needed a way to improve how they handled subscriptions for the events they hosted.

The old way was with multiple people using different places to track the subscriptions, which led to a lot of misleading information. The project ended up helping them to centralize not only the subscriptions but also have a more precise estimation of participants, integration with payment service, and complete financial reports.

• Built the application from scratch, starting with gathering the requirements and designs on Figma and ending on the hosting on Digital Ocean (back end and database) and Vercel (front end);
• Built a GraphQL API connecting to a PostgreSQL database with dynamic SQL queries based on the GraphQL query;
• Implemented dynamic reports for the manager to select the exact information they wanted to see and generate a PDF file with that information;
• Built a CMS that allowed the event manager to create a custom page with a link and send it to their WhatsApp groups for sign-ups;
• Integrated the platform with a payment service called GalaxPay, making it easy to implement secure PIX (Brazilian instant payment method) transactions.