Data and computing projects

Context

Much of my technical growth has happened through project-based work across university, internship, and team settings.

The projects covered very different domains, from machine learning and NLP to embedded audio processing and FPGA implementation.

What I did

• worked on copy-paste augmentation and Mask R-CNN experimentation
• contributed to predictive modelling for community renewable energy data
• used NLP and LLM-related approaches across different tasks
• handled data wrangling and linkage work with privacy considerations
• built real-time audio signal processing on STM32
• implemented a smart-home controller on FPGA
• carried out clustering analysis and comparison work

Results

• built working prototypes across both software-heavy and hardware-constrained contexts
• took experiments beyond implementation into evaluation, comparison, and explanation
• gained practical experience with privacy-aware data handling rather than only classroom-style modelling
• developed a broader technical range than a single-language or single-stack project history would suggest

What this shows

These projects taught me not only tooling, but range. I learned how to move between different technical contexts, understand what a project is actually trying to achieve, and turn methods into something testable and explainable.

The same pattern showed up again in the Accenture Tech Bootcamp, where our team placed first after designing and deploying a platform on AWS using Amplify and serverless technology. It was a good reminder that technical work becomes more valuable when it can be communicated, shipped, and improved collaboratively.