Roza Dler

Machine Learning Engineer

rozadler.rd@gmail.com

I’m a Machine Learning and AI Engineer with a strong passion for research, innovation, and applying AI to real-world challenges. My journey began in cross-platform mobile application development, where I enjoyed building functional, user-focused solutions. I was then recipient of the Chevening Awards and pursued MSc in Artificial Intelligence at the University of Surrey in the UK, and graduated with distinction.

During my master’s, my dissertation focused on generalizing biomedical image analysis and classification using multimodal, foundational, language and vision models. This project not only strengthened my technical and research skills but also deepened my interest in the intersection of AI and healthcare. Beyond healthcare, I’ve worked on projects ranging from knife classification models to support the justice system to time-series and audio recording analysis for heart murmur detection, reflecting my passion for applying AI across diverse fields.

I’m also an alumna of the Iraqi Young Leaders Exchange Program (IYLEP) and earned my BSc in Information Technology from the American University of Iraq, Sulaimani, where I minored in English Literature—both supported through competitive scholarships. Whether I’m cleaning and analyzing data, building machine learning models, or contributing to inclusive, multidisciplinary teams, I’m driven by a love for continuous learning, innovation, and a commitment to impactful work.

Experience

Dec 2022 - Nov 2023

Lead Developer & Mobile Application DeveloperTo U Delivery Service

  • Led the entire development process, from requirements gathering and design to deployment and maintenance.
  • Designed, developed and deployed two cross-platform applications, Manager and Driver, using Flutter and Firebase.
  • Led and collaborated with a cross-functional team (UX/UI Designer and Web Developer) to design and develop the Admin dashboard using Vue.js.
  • Assessed and prioritised feature requests by stakeholders, evaluating feasibility, impact, and business value.
FlutterDartFirebaseVue.jsMaps APIGCPFigma

Sep 2019 - Dec 2022

Translator and InterpreterPeople’s Development Organization and Deir Mariam Al-Adra Monastery

  • Interpreted live communications between European directors, producers, and local actors and staff during all phases of theatrical productions, including rehearsals, exercises, and workshops for the Sabunkaran Theatre Group Productions.
  • Translated key written materials such as scripts, video documentaries, and production guidelines, ensuring precise and culturally appropriate language.
  • Provided simultaneous interpretation for PDO’s workshop on Mediation and Conflict Resolution for social workers and carers, collaborating with organisers to adapt the curriculum based on participant feedback.

Feb 2019 - Jan 2020

Research Assistant and CoordinatorCenter for Gender and Development Studies (CGDS)

  • Played a pivotal role in the EU-funded project “Enhancing Education, Developing Community, and Promoting Visibility to Effect Gender Equity in Iraq and the MENA Region”, which aimed to reform educational curricula through revision of primary and middle school textbooks.
  • Conducted data analysis and compiled detailed reports and visual summaries, contributing to recommendations for the Ministry of Education to advance inclusive and equitable education.

Projects

A picture showcasing the projectA picture showcasing the project

MSc Dissertation Project

My MSc dissertation, Rethinking biomedical Image Classification with Transformers and vision-Language Foundation models, aimed to generalise biomedical image classification regardless of classification tasks or image volume and modality.

PythonPytorchWandBGitLabDocker
A picture showcasing the project

CNN For Scene Recognition

The aim of this project was to use Convolutional Neural Networks (CNNs) and Transfer Learning for scene recognition. A resnet34-based network was trained to classify 40 different scenes.

PythonPytorchTensorBoard
A picture showcasing the project

Vehicle Re-Identification

This project aimed to develop a vehicle re-identification model using deep learning. The model was trained on the VehicleID dataset.

PythonPytorchTensorBoard
A picture showcasing the project

Heart Murmur Detection

A Joint Learning Model for heart murmur detection through Phonocardiogram Recordings and time series patient records

PythonPytorch
A picture showcasing the project

Yellow

Yellow, A mood tracking application, was my final year project at university for my BSc. It allows users to track their mood and mental health over time. Users get to keep a diary of their mood fluctuations and be able to have a visual representation of their moods through colors, charts, and calendar view.

FlutterDartFirebaseFigma