Leonardo Muñoz

•Developed and optimized global software solutions to enhance fleet vehicle sales automation, contributing to $26 billion in annual revenue. •Developed comprehensive unit and integration tests using xUnit, Jest, Mocha, ensuring code quality and reducing production incidents. •Implemented CI/CD pipelines (GitHub Actions, Azure DevOps) to automate deployment and improve release cycles. •Engineered front-end solutions across .NET, MERN, and Java tech stacks, ensuring seamless alignment with UX/UI team specifications. •Developed RESTful and GraphQL APIs, enhancing integration with frontend applications and external services. •Designed and implemented scalable micro-services using .NET Core, Docker, and Kubernetes, improving system modularity and maintainability. •Managed full lifecycle ownership of two critical enterprise applications, driving key feature development, maintenance, and system performance improvements.

•Wrote scripts in Bash and PowerShell to aggregate data from Windows and Linux devices connected to the White Sands Test Facility's network with the goal of ensuring all devices were following baseline security measures. • The PowerShell and Bash scripts enabled the IT Security team to decrease the time it took to perform compliance audits from 2-3 hours per device, to 5-10 minutes for a batch of up to 100 devices. • Created an object-oriented data mining and analysis program from scratch using Python. • The Python tool was reworked in JavaScript, HTML, and CSS to provide more flexibility of use within NASA’s White Sands Testing Facility’s highly restrictive environment.

Developed a CAN Bus Visualizer that was capable of identifying the topography of a CAN Bus using the J1939 protocol and tap into the packets being sent across the network. The tool automatically decodes the packets and allows the user to manipulate, replay, and send custom packets across the CAN bus.

•Assisted in training and fine-tuning Natural Language Understanding models using BERT for text classification tasks, improving intent recognition accuracy. •Worked with TensorFlow and PyTorch to experiment with CNN based image classification models, optimizing hyper parameters to enhance model performance. •Developed data pipelines for preprocessing and feature extraction, which enabled smoother training and evaluation of ML models on large datasets.

- TI Summer Project 2020. - Used the TI RSLK MAX as a base to build a prototype robot. - Built a robot that autonomously drives itself capturing data on air quality. If air quality ever fell below-set standards, it would alert the user of such occurrence.