Data Scientist

• Aug 2023 - Present

Data Scientist

• Sep 2022 - Jul 2023

As a Data Scientist at FieldBox, I am responsible for developing, deploying and troubleshooting machine learning models, monitoring and analysis of the models’ performance and their impact, and the maintenance of data pipelines for the smooth functioning of ML products. I have also been fortunate to contribute to the internal tech stack, being a part of the AI engineering team. Some of the topics that I am working on: 1. Building pipelines for delivering real-time alerts from time… Show more As a Data Scientist at FieldBox, I am responsible for developing, deploying and troubleshooting machine learning models, monitoring and analysis of the models’ performance and their impact, and the maintenance of data pipelines for the smooth functioning of ML products. I have also been fortunate to contribute to the internal tech stack, being a part of the AI engineering team. Some of the topics that I am working on: 1. Building pipelines for delivering real-time alerts from time series predictive models - Deployed models trained on structured, longitudinal data using Kedro and airflow pipelines - Performed EDA, visualisations, prototyping, and conversion of data science scripts to pipelines based on the business requirements - Fixing systemic and end-to-end issues in the data pipelines 2. Contributing to the development of internal tools - Developed a CI-friendly programming connector for interacting with Microsoft SQL databases using a Python-based interface replacing SQLAlchemy - Enabled the logging of intermediate alerts at different stages of data and modelling pipeline to Sentry Show less As a Data Scientist at FieldBox, I am responsible for developing, deploying and troubleshooting machine learning models, monitoring and analysis of the models’ performance and their impact, and the maintenance of data pipelines for the smooth functioning of ML products. I have also been fortunate to contribute to the internal tech stack, being a part of the AI engineering team. Some of the topics that I am working on: 1. Building pipelines for delivering real-time alerts from time… Show more As a Data Scientist at FieldBox, I am responsible for developing, deploying and troubleshooting machine learning models, monitoring and analysis of the models’ performance and their impact, and the maintenance of data pipelines for the smooth functioning of ML products. I have also been fortunate to contribute to the internal tech stack, being a part of the AI engineering team. Some of the topics that I am working on: 1. Building pipelines for delivering real-time alerts from time series predictive models - Deployed models trained on structured, longitudinal data using Kedro and airflow pipelines - Performed EDA, visualisations, prototyping, and conversion of data science scripts to pipelines based on the business requirements - Fixing systemic and end-to-end issues in the data pipelines 2. Contributing to the development of internal tools - Developed a CI-friendly programming connector for interacting with Microsoft SQL databases using a Python-based interface replacing SQLAlchemy - Enabled the logging of intermediate alerts at different stages of data and modelling pipeline to Sentry Show less

Research Associate

• Sep 2019 - Aug 2022

AI researcher applying computer vision for the advancement of human vision. - Development of deep learning computer vision models for ophthalmological tasks like the detection of ocular ailments and the prediction of systemic features - Applying upstream advances in computer vision like GANs or vision transformers for solving issues in the ocular domain - Automation and the usage of AI for the reduction of manual effort Selected publications: 1. Automatic segmentation of… Show more AI researcher applying computer vision for the advancement of human vision. - Development of deep learning computer vision models for ophthalmological tasks like the detection of ocular ailments and the prediction of systemic features - Applying upstream advances in computer vision like GANs or vision transformers for solving issues in the ocular domain - Automation and the usage of AI for the reduction of manual effort Selected publications: 1. Automatic segmentation of corneal deposits from corneal stromal dystrophy images via deep learning. Computers in Biology and Medicine, October 2021 https://doi.org/10.1016/j.compbiomed.2021.104675 2. Deep learning algorithms for automatic detection of pterygium using anterior segment photographs from slit-lamp and hand-held cameras. British Journal of Ophthalmology, July 2021 http://dx.doi.org/10.1136/bjophthalmol-2021-318866 3. Prediction of systemic biomarkers from retinal photographs: development and validation of deep-learning algorithms. Lancet Digital Health, October 2020 https://doi.org/10.1016/S2589-7500(20)30216-8 Show less AI researcher applying computer vision for the advancement of human vision. - Development of deep learning computer vision models for ophthalmological tasks like the detection of ocular ailments and the prediction of systemic features - Applying upstream advances in computer vision like GANs or vision transformers for solving issues in the ocular domain - Automation and the usage of AI for the reduction of manual effort Selected publications: 1. Automatic segmentation of… Show more AI researcher applying computer vision for the advancement of human vision. - Development of deep learning computer vision models for ophthalmological tasks like the detection of ocular ailments and the prediction of systemic features - Applying upstream advances in computer vision like GANs or vision transformers for solving issues in the ocular domain - Automation and the usage of AI for the reduction of manual effort Selected publications: 1. Automatic segmentation of corneal deposits from corneal stromal dystrophy images via deep learning. Computers in Biology and Medicine, October 2021 https://doi.org/10.1016/j.compbiomed.2021.104675 2. Deep learning algorithms for automatic detection of pterygium using anterior segment photographs from slit-lamp and hand-held cameras. British Journal of Ophthalmology, July 2021 http://dx.doi.org/10.1136/bjophthalmol-2021-318866 3. Prediction of systemic biomarkers from retinal photographs: development and validation of deep-learning algorithms. Lancet Digital Health, October 2020 https://doi.org/10.1016/S2589-7500(20)30216-8 Show less

Research Intern

• Jan 2019 - Sep 2019

Prognosis and detection of diabetic retinopathy and related pathologies from high-resolution fundus images using deep learning Prognosis and detection of diabetic retinopathy and related pathologies from high-resolution fundus images using deep learning