MIA - Luis De Trejo

Role

Software Developer

Timeline

2022 - 2024

Technologies

Django, Django REST Framework, React Native, Celery, RabbitMQ, Stripe, AI/ML (Deep Learning, Transfer Learning, CNNs), Firebase Cloud Messaging, Firebase Analytics

Project Overview

My Image Advisory is a cutting-edge mobile application that harnesses the power of artificial intelligence to deliver personalized fashion recommendations. As a key developer, I played a significant role in designing and implementing both the backend and mobile app

My work focused on creating a robust, scalable architecture that supports advanced AI functionalities, ensuring a seamless and enjoyable user experience.

Key Contributions

Backend Development

Developed a scalable backend using Django and Django REST Framework, adhering to best practices for web services and ensuring robust and efficient data handling.

AI Algorithm Enhancement

Retrained existing AI algorithms, including body shape classification, face shape classification, and colorimetry classification, utilizing modern deep learning techniques and transfer learning to enhance recommendation accuracy.

Mobile App Development

Contributed extensively to the React Native mobile app, implementing a modular architecture, integrating TanStack Query for data management, and optimizing user interfaces to improve engagement.

Background Task Management

Implemented Celery and RabbitMQ for efficient background task processing, enhancing app performance and reliability.

Payment Integration

Successfully integrated Stripe for managing subscription plans and processing payments, ensuring a smooth and secure user experience throughout the payment flow.

User Engagement

Implemented Firebase Cloud Messaging for push notifications to keep users engaged, along with Firebase Analytics to gather actionable insights into user behavior.

Challenges and Solutions

One of the primary challenges was seamlessly integrating sophisticated AI algorithms into the mobile app without compromising performance. This was addressed by optimizing server-side processing and streamlining data transfer between the server and client.

Another challenge involved integrating Stripe to handle various edge cases in subscription management across multiple platforms. This was resolved by implementing a robust order flow that guarantees successful transactions and renewals.

Results and Impact

Although specific user retention statistics cannot be disclosed due to client confidentiality, the redesigned platform saw a marked improvement in user satisfaction thanks to the accurate and personalized fashion recommendations powered by AI. The iterative process of refining the AI models and optimizing the app architecture resulted in a reliable and engaging user experience.

My Image Advisory