Hello, I'm Salviya.
Hello! My name is Salviya (Joshua) Balami, and I’m currently a student at Caltech studying Computer Science and Business, Economics, and Management. I’ve developed a strong foundation and growing interest in machine learning, robotics, and full-stack development, with experience in technologies that help solve real-world problems through creative and technical solutions.
I'm particularly passionate about applying these skills toward the development of autonomous vehicles, and I’m actively looking to pursue a career in this space — especially at the intersection of machine learning, perception, and control systems.
This site highlights my academic journey, projects I've built, and the experiences that have shaped my path so far. Whether you're a potential collaborator, recruiter, or simply curious, I hope you find something here that resonates with or inspires you. Feel free to explore my past projects and research, check out my GitHub, and connect with me on LinkedIn. I’m always open to conversations, ideas, and new opportunities!
Before college, I attended the Math & Science Academy in Woodbury, Minnesota — the #1 ranked high school in the state — where I completed both middle and high school. During my time there, I explored a range of advanced problem-solving techniques and took rigorous courses in calculus, introductory computer science, and the core sciences, including biology, chemistry, and physics. I also built a strong foundation in Python and Java while developing my analytical thinking and curiosity for STEM.
I currently attend Caltech, where I study a range of courses in programming fundamentals, software design, and web development. Through this coursework, I’ve gained hands-on experience and growing proficiency in languages such as Python, Java, C, HTML & CSS, and JavaScript. I'm particularly passionate about applying machine learning and computer vision to the development of autonomous vehicles, with a focus on enhancing their awareness and steering capabilities.
During my time at Caltech, I’ve contributed to several hands-on projects that reflect both my technical range and interests. I built BeaverMaps, a Java-based route-finding app that uses real-world map data and optimized search algorithms to navigate Caltech’s campus. I created Triviatron, a full-stack trivia game using Node.js and JavaScript with streak-based scoring and dynamic UI updates. In C, in collaboration with my roomates, we developed a physics-based Doodle Jump engine compiled to Emscripten, with real-time collision detection and game state logic. Most recently, I launched Baseline Threads, an NBA-themed e-commerce site that displays real-time jersey data using API integration and client-side filtering to enhance the browsing experience.
I’m currently conducting research at the University of Minnesota as part of the NSF REU program, Human-Centered Computing for Social Good. My work centers on developing a cost-effective teleoperative arm for the Boston Dynamics SPOT robot to enhance user accessibility and control speed. In parallel, I’m implementing an imitation learning framework to enable SPOT to perform real-world pick-and-place tasks by learning from human demonstrations.