● Focus on stabilizing the NDVI metric across different weather conditions by conducting exploratory data analysis and developing Machine Learning algorithms.
● Analyze existing data to derive metrics on how the health index for a given field changes through time with respect to fertilizer application and plant growth.
● Developed a new algorithm, the generation of differentiated field-maps between 2 operations for a field, showing the progress of the field during a period. (used Machine Learning algorithms)
● Develop an automated-dynamic customer's PDF report generation for the aforementioned project, using ReportLab and matplotlib methods.
● Incorporate the entire procedure with the Slack chat platform by building a Slack API Bot.
● Used Object Oriented Programming (OOP) throughout this project and contributed on several existing repositories of prior engineers.

Member of the AUEB Natural Language Processing (NLP) Group. Conducting research on Medical Image Captioning (Diagnostic Captioning) on Biomedical datasets, using uni-modal, cross-modal as well as multi-modal Deep Learning models.
More precisely, Computer Vision (CV) approaches like Convolutional Neural Networks (CNNs), Transformers (Vision and non), Neural Retrieval algorithms, were employed throughout this research.
In addition, I attend several group meetings, in which we analyse recent research papers in the field of Natural Language Processing (NLP).

During my thesis research, as a member of AUEB NLP-Group (in particular the sub-group which conducts research on Biomedical fields) , I took part in the annual ImageCLEFmedical Caption Task campaign. Being the main driver of this team for diagnostic captioning, I developed deep learning systems that ranked 2nd in the primary evaluation measure and 1st in secondary evaluation measure, among 10 teams, in the Caption Prediction Task. (For more information: https://www.imageclef.org/2022/medical/caption)

Our paper: http://ceur-ws.org/Vol-3180/paper-101.pdf