Hi, I’m Salvador Sanchez — I build software that gets out of its own way.
CS student at Wayne State. I build software that gets out of its own way, from ML backend to interface. Currently in Dr. Zhu's lab researching adversarial robustness in embodied AI.
A little more context.
I started writing code originally due to my love for sports. I was the kid who was obsessed with stats, whether it was baseball, soccer, or football. I was always curious about what differentiatied the greats from the rest, and that question eventually led to me create Litix.
Along the way I fell in love with parts of software that aren't related to sports. Primarily machine learning at first, designing and training models to predict more than just player performance. I got pulled into the logic behind backend systems and ML models, but the part that surprised me was the design, realizing that a good interface isn't just decoration, it is making the software as intuitive and frictionless as possible, not making the user think about how to use it.
Currently I am a research assistant in Dr. Zhu's lab at Wayne State, where we are working on adversarial robustness in embodied AI. We are exploring how to make AI systems more robust to adversarial attacks, which is crucial for their safe deployment in the real world. It's a fascinating area of research that combines my interests in machine learning and software design, and I'm excited to see where it takes me.
Things I’ve built.
-
/01
2025Litix - Fantasy Football Analytics Platform ↗
A fantasy football management and analysis platform powered by Spring Boot and Google Vertex AI (Gemini). Using a ML model I built, trained, and deployed as an API, it provides users with data-driven insights and tools to optimize their fantasy football teams, including player performance predictions, trade recommendations, and lineup optimizations.
-
/02
2026MoodLens ↗
A real-time emotion and stress monitoring desktop app that uses your webcam to track facial expressions, gaze, and stress levels — then adapts your environment to help you stay focused and calm. Dashboard illustrates stress paterns based on time, day, and app usage to help users understand their stressors and manage them proactively. Built using PyQT for the interface, OpenCV/ MediaPipe for computer vision, and a custom-built scikit-learn model to predict user stress prior to it happening.
-
/03
2026Pomodoro Clock ↗
A pomodoro clock app with a dashboard, and tasks list. The app helps users study more effectively by breaking work into intervals, tracking progress, and providing insights on productivity patterns.
Where I’ve worked.
-
Oct 2025 — PresentOperations StrategistOrange Sparkle Ball · Technology Focus
Lead QA testing for autonomous robot hailing and logistics platform, validating endpoints, user flow, and system reliability. Collaborated with developers on a user platform that allows users to schedule a pick-up from autonomous robot partners. Supported cloud deployment, monitoring, and workflows using AWS.
-
Winter 2026 — PresentUndergraduate Research AssistantEmbodied AI Lab · Wayne State University
Working with Prof. Zhu on investigating adversarial attack vulnerabilities in Vision-Language Model (VLM) and Vision-Language Action (VLA) based embodied AI agents.
-
Oct 2024 — Oct 2025Software EngineerBackyard Bats
Helped with the devolpment and design for a website that allows users to buy and design their own custom bats. Helped with creating logic for the product customizer, as well as the user interface and experience of the website.
Let’s talk.
I’m always happy to hear from people working on interesting problems — recruiters, fellow students, or anyone who shares the same interests. The fastest way to reach me is email.