The old website was built with plain HTML.
At the start of my internship, the Centre of Forensic Science's technology stack was extremely outdated. Despite serving thousands of forensics investigators, the user interface was poorly designed and difficult to navigate. The process for updating content was highly inefficient, often taking several days.
Whenever a change needed to be made, the Business department sent an email to the IT department requesting the change. After the task was delegated to IT staff, they would go into the source code to edit the HTML directly. There would then be a back and forth between the IT and Business departments about fixing minor issues or misinterpreting what needed to be changed. My job as a Software Intern was to figure out a solution to these problems.
I migrated the entire website to NextJS
My co-intern and I rebuilt the entire website in NextJS. We redesigned the user interface while adhering to Ontario government standards, making it easier for forensics clients to locate relevant resources. I set up a headless CMS using Stackbit and built a CI/CD pipeline using Azure Static Web Apps. This enabled the Business department to update and publish website content without IT support, reducing website maintenance times by over 10x.
AI Chatbot & Internationalized Routing
Aside from migrating the old website, I prototyped an AI chatbot that used natural language processing to answer daily inquiries from 4,000+ clients. Technologies used included RASA, OpenAI API, and Docker. I also helped implement a French version of the website using internationalized routing in NextJS.