Microchip | Vancouver, BC | Jan. 2025 – Present

• Developed a reference repository to assist developers at NASA in integrating external libraries compatible with SoC’s Linux OS, streamlining third-party library adoption for spaceflight applications.
• Automated external library integration by designing a Jenkins pipeline to build library artifacts, developing Bash scripts, and enabling compatibility with an Ubuntu-based Docker container, reducing manual effort and streamlining deployment.
• Followed Agile methodology with regular sprints and backlog management. Adhered to DevOps practices for CI/CD, ensuring automated builds, testing, and deployments.

The Ubyssey Publications Society | Vancouver, BC | Mar. 2024 – Present

• Contributed to developing a Content Management System for Ubyssey, enabling editors to create e-newspapers efficiently as a lead developer. The website is read by 120,000 unique users every month.
• Developed a responsive events page using React along with SASS, HTML, and Django. This page became the second highest viewed on the website, achieving over 2,000 daily views.
• Implemented automated cross-browser regression testing with Selenium in Docker, integrating it into a CI/CD pipeline via GitHub Actions to prevent critical bugs from reaching production.
• Improved website performance by 20%, measured by Google Lighthouse, by reducing the loading time of the website.

• Designed a robust traffic sign recognition pipeline for autonomous vehicles using an Intel RealSense camera and OpenCV.
• Developed and trained a Convolutional Neural Network (CNN) using TensorFlow, achieving 99.4% validation accuracy.
• Integrated image preprocessing and machine learning nodes through ROS2 messaging for real-time traffic sign evaluation.
• GitHub: Project Link

• Developed a web app to help store owners track customer behavior by leveraging YoloV8’s computer vision system.
• Created a responsive frontend using React, integrating it with Flask for real-time analytics.
• Implemented a MongoDB data pipeline for computing insights on customer movement and store traffic trends.
• GitHub: Project Link

• Built a deep learning accelerator for an embedded Nios II system to classify handwritten digits from the MNIST dataset.
• Interfaced hardware and software communication, working with off-chip SDRAM for efficient processing.

• Developed an interactive Blackjack game for Android using Java and XML, enhancing user engagement.

• Conducted extensive testing across different device form factors to ensure responsiveness and backward compatibility.

• GitHub: Project Link

  Engineering Placement Prediction

• Developed an ML model to forecast university student job placements, achieving 88.7% accuracy.

• Conducted data preprocessing, feature engineering, and applied classification algorithms to improve model accuracy.

• Fine-tuned the model using hyperparameter optimization with grid search, reducing prediction error by 5%.

• GitHub: Project Link

• Built a client-server architecture with socket programming for communication between sensors and an IoT controller.
• Utilized multi-threading for concurrent client processing, boosting performance by 40%.

• Implemented a simple RISC architecture supporting ARM instructions.
• Designed ALU, Shifter, Registers, and integrated them into the datapath with an FSM for control.
• Verified synthesis by compiling the design on Quartus and testing RTL synthesis with gate-level testing.
• GitHub: Project Link