Gustavo – Python, React, Django

It's a visual discovery platform where users find, save, and organize images and videos to collect ideas and inspiration.

• Implemented critical, well-tested features in a social media platform used by over 500 million users.
• Reduced the loading time of heavy GIF images from 450 ms to 98 ms.
• Built a new tracking and logging system for uploads.
• Integrated internal queues to run heavy processes in the background.
• Implemented new routes in the ATS to connect with Zendesk.

It's an internal platform that centralizes, retrieves, and creates ticket information from Zendesk and Jira issues. All information is gathered in one place to support employees, managers, and the support team.

• Created a platform used by more than 4,000 employees worldwide.
• Reduced annual AWS costs by 20% by the end of the project.
• Integrated the platform with several external applications, including Zendesk, Jira, and Workday.
• Delivered the project three weeks ahead of schedule due to strong organization.

It's a product that enables marketing agencies to batch-upload campaign creatives and distribute them across multiple advertising vehicles simultaneously.

• Led a 3-member team (1 mid-level, 2 junior engineers).
• Built the project end-to-end, from system design to architectural decisions.
• Delivered a system that reduced processing time from 3 days to 15 minutes.
• Translated business requirements into well-tested production features.

It's a platform that streamlines the approval process for marketing assets (videos, GIFs, images) and is used by marketing agencies, clients, and campaign managers.

• Implemented a new process for managing campaign financial data, reducing manual errors by 15% and accelerating review and approval by 40%.
• Created asset processing pipelines (images, videos, GIFs) to analyze, compress, and transmit content.
• Built media players from scratch for all asset types to provide better, controlled visualization.
• Implemented a chat system that allows campaign managers to communicate directly with clients.

It is a marketing campaign management project featuring an internally built spreadsheet system for managing budgets and expenses. It also includes history tracking, spreadsheet checkpoints, and integration with advertising platforms.

• Created the project from scratch, making decisions on architecture, technologies, and delivery model.
• Integrated an internal spreadsheet system to enable two-way compatibility with existing Excel spreadsheets.
• Built a checkpoint and rollback system for all spreadsheet states.
• Implemented information sharing between users and reduced manual errors by over 60%.

It is a portal for health insurance plan beneficiaries, where users can schedule appointments, have teleconsultations, view their medical history, and access their membership card information.

• Primarily responsible for the front end of the application, including nearly all development and production configuration.
• Experience working with very large codebases, following well-defined architectures and clean code practices.
• Implementation of multi-step forms focused on appointment scheduling.
• Active participation in decision-making related to architecture, design, and user flow.

It was an application focused on mental health, with features such as meditation, thought sharing, and connecting users with sponsors to support addiction recovery.

• Developed a cross-platform application focused on mental health to support people.
• Integrated external platforms such as Spotify to fetch music for meditation sessions.
• Designed and built a journaling system from scratch for recording thoughts.
• Implemented separate user and sponsor roles, each with their own specific functionalities.

It is a dashboard and factory valve failure prediction project. It uses machine learning models and big data to perform predictions and display historical data.

• Actively participated in meetings with end clients, consistently keeping the customer perspective in focus.
• Created simple machine learning models using Python.
• Developed the entire front end using vanilla JavaScript, HTML, and CSS.
• Integrated predictive models with the server via REST APIs.