CARE: Building an AI-Driven Wildlife Conservation Platform

One of the most rewarding projects I’ve been heavily involved in was during my Computer Science capstone course at the University of Auckland. I stood at the helm of a talented 6-member team, leading the development of CARE: a specialized Web Platform built for wildlife conservation.

CARE was more than an academic exercise—it served a real-world purpose. We aimed to create a web-based portal that could employ AI-based image recognition to effectively track predator populations, contributing substantially to New Zealand biodiversity protection.

Structuring a Complex Project

Our mission was ambitious, and my dual role as the team lead and full-stack developer meant orchestrating clear technical direction and empowering every member.

Here’s how we structured our teamwork:

• Agile Development: I introduced our team to strict Agile methodologies. We held regular sprint planning sessions, maintained continuous task-tracking through GitHub Projects, and synchronized our efforts via frequent stand-ups.
• Mentorship: Not everyone on the team had strong experience with certain backend tools, particularly Sequelize. I solved this by preparing “quick how-to” documentation and conducting brief, intensive 1-on-1 support sessions. This helped close the knowledge gaps, boosted morale, and kept the timeline tight.
• Cross-Functional Syncing: Part of this build involved deeply integrating the platform with AI image processing engines provided by PhD researchers at the university. This required maintaining very open bridges of communication between frontend, backend, and the specialized AI logic.

Technical Implementation

From an architectural standpoint, CARE was designed as a modern JavaScript-stack web application:

• Frontend: The user interface was built using a responsive React architecture wrapped in clean HTML/CSS. This was critical to ensure researchers interacting with the dashboard found it accessible and user-centric.
• Backend: I worked heavily on constructing the core backend APIs utilizing Node.js and Express.js, wiring it to a functional relational database powered by Sequelize.

Outcomes and Takeaways

Delivered flawlessly on time, CARE received outstanding feedback. Not only was it praised for its robust core functionality of managing image-based pest detection workflows, but its clean user journey also turned heads.

More than any lines of code written, leading CARE solidified the importance of clear, empathetic communication and the ability to adapt as priorities pivot. Combining robust technical execution with a hyper-focused user/researcher mindset is something I now strive for in every team I am part of.