Vision Science & Machine Learning Research Intern

Location: Singapore, SG

Vision Science & Machine Learning Research Intern

Brand

EssilorLuxottica

Location

Singapore, SG

#job-location.job-location-inline {

display: inline;

}

WHO WE ARE

We are EssilorLuxottica, a global leader in the design, manufacture and distribution of ophthalmic lenses, frames and sunglasses. The Company brings together the complementary expertise of two industry pioneers, one in advanced lens technologies and the other in the craftsmanship of iconic eyewear, to create a vertically integrated business that is uniquely positioned to address the world’s evolving vision needs and the global demand of a growing eyewear industry.

With over 200,000 dedicated employees in 150 countries driving our iconic brands, our people are creative, entrepreneurial and celebrated for their unique perspectives and individuality. Committed to vision, we enable people to “see more and be more” thanks to our innovative designs and lens technologies, exceptional quality and cutting-edge processing methods. Every day we impact the lives of millions by changing the way people see the world.

JOB SCOPE AND MAIN RESPONSIBILITIES

Your mission will be to support the Applied Myopia Research R&D team in developing and validating prediction models for myopia progression and treatment outcomes, preferably using machine learning techniques within the Applied Myopia Research domain.

You will be involved in data quality checks and exploratory data analysis, data preparation, development and testing of machine‑learning–based prediction models, and monitoring model performance using appropriate validation metrics. You will also participate in documenting methodologies and results, and in communication of findings and technical discussions with the project team.

The expected results include reliable data models, well-documented methods, and clear summaries of findings. The main challenges would be learning and applying appropriate machine learning techniques for prediction models for myopia progression and control, and managing project timelines. Success requires scientific curiosity, attention to detail, initiative, and effective teamwork.

AREAS OF RESPONSIBILITIES AND RELATED ACTIVITIES

Prediction Model Development and Evaluation

Develop and evaluate statistical and machine-learning models for predicting myopia progression and treatment outcomes.

Apply appropriate supervised learning techniques to longitudinal ophthalmic and clinical datasets.

Compare different modelling approaches and identify suitable models for specific prediction objectives.

Tune model parameters and optimize model performance while considering interpretability and clinical relevance.

Evaluate model performance using appropriate metrics, such as RMSE, MAE, R², calibration, discrimination, and prediction error, as appropriate to the modelling task.

Contribute to the development of reproducible modelling pipelines.

Model Validation and Performance Monitoring

Implement appropriate training, validation, and testing strategies for predictive models.

Perform cross-validation and/or external validation where appropriate.

Assess model generalizability, robustness, calibration, and potential overfitting.

Compare model predictions against observed longitudinal outcomes.

Assist in monitoring model performance across relevant subgroups, such as different ages, baseline refractive errors, or study populations.

Documentation and Reproducible Research

Document data-processing procedures, modelling methodologies, assumptions, parameters, and validation approaches.

Maintain well-organized and reproducible analysis scripts and workflows.

Prepare tables, figures, technical summaries, and other analysis outputs.

Contribute to internal technical reports, presentations, and potentially scientific publications or conference materials.

Scientific Analysis and Interpretation

Interpret modelling results in collaboration with scientists and researchers in the Applied Myopia Research R&D team.

Relate statistical and machine-learning findings to the biological and clinical aspects of myopia progression.

Identify limitations, assumptions, and potential sources of bias in prediction models.

Contribute to scientific discussions regarding model development and application.

Communication and Team Collaboration

Present analytical findings and model performance to the project team.

Communicate technical concepts and results clearly to colleagues with different levels of data-science expertise.

Participate in technical discussions and contribute ideas for improving modelling approaches.

Work collaboratively with data scientists, clinicians, researchers, and other R&D team members.

KEY REQUIREMENTS

Currently pursuing a Master's degree in Big data engineering, data science, computer science, biostatistics & biomedical data, or a related discipline.

Able to commit for a minimum of 6 Months (Jan - Jun/Jul, 2027)

Proficiency in R and/or Python for data analysis and statistical/machine-learning modelling.

Familiarity with data manipulation, visualization, statistical modelling, and machine-learning methods.

Experience with packages such as tidyverse, ggplot2, tidymodels/caret (R) or pandas, NumPy, scikit-learn, and matplotlib (Python) is preferred.

Familiarity with machine-learning methods such as linear and generalized linear models, regularization, tree-based methods, random forests, gradient boosting, and model validation is an advantage.

Experience with Git/GitHub, R Markdown/Quarto, or Jupyter for reproducible analysis is an advantage.

Experience or interest in longitudinal data analysis, predictive modelling, and/or biomedical data science is highly desirable. Knowledge of mixed-effects models, repeated-measures data, or time-dependent prediction is an advantage.

Reliable, autonomous, precise, good organization & communication skills

Positive and proactive attitude

Recruiting process

Our recruitment process may