About me:

With a passion for sustainability and a deep understanding of Life Sciences , I am committed to using data to create a better future for everyone. Whether it's by identifying new areas of innovation, developing predictive models to optimize resource use, or using data-driven insights to promote social and environmental responsibility, I am always looking for new ways to make a positive impact through data science.

Python

Project 1: Effects of climate change on plant disease enviornmental parameters (Master's Thesis)

Links:   Interactive app (python Dash)   |  Results (jupyter notebook)   |   Code (github)

[WORK IN PROGRESS]

• access publicly available weather data for Germany (1950 - present) trough a REST API
• apply different wetness models on the datasets
• create a python Dash to communicate the results of the study (see in "Links")

Project 2: Automatizing imports from customers' agricultural enterprise [private repository]

• importing different types of files (shp, csv, iso-xml etc.) and converting them into the internal structure
• improving the internal structure for storing and accessing data
• creating a database and matching imported objects trough feature engineering
• keeping the database up to date while improving the matching efficacy

Project 3: Web scraping and API accessing FAO EPPO

• web scrape a table with crop names (in English and latin) from FAO website
• get the EPPO code from the API using the latin name
• merge all the data in a dataframe, export it as a csv

R

Project 4: Replicating the results of a published research (University project)

Links: Results (HTML page)   |   Code (github repository)

• replicating the results of: Effects of chronic exposure to thiamethoxam on larvae of the hoverfly Eristalis tenax (Diptera, Syrphidae) - a study on the effect of insecticides on pollinators
• project made in collaboration with 3 classmates from PlantHealth Master's program
• statistical methods applied: ANOVA, Shapiro, Kruskal-Wallis
• visualisation methods: Kaplan-Meier curves, elipses, Heatmap , Correlation circle

Fig. 2.3: Correlation circle between the 7 different behaviours. Variables grouped together are corelated, while the arrows closer to the circle are better explained by the two dimensions of the PCA. (F,W correlated; S, GR correlated; M-PR negatively correlated but PR is not well explained by the two dimensions) Keys: S - stationary, GR - grooming, W - walking, F - flying, PR - probing, N - feeding(nutrition), M - moving

SQL

Project 5: Create and maintain a database for agriculture products, processes and crops (SQLite) [private repository]

• create a relational database that would correspond to present and future needs
• import and adapt the data from data formats currently in use (csv, json) while setting the right primary and foreign keys for tables
• implement the database in production (Rshiny app) without disturbing the current functionalities of the product

Additional information

Courses and certificates

• Understanding TypeScript - 2023

• TypeScript for Professionals -2022 Edition

• SQL for Datascience

• Applied Machine Learning in Python

• Applied Plotting, Charting & Data Representation in Python

• Statistics with Python

• Google Data Analytics Professional Certificate - [WORK IN PROGRESS]

Areas of improvements

It takes a lot of strength to admit your weaknesses. However, admitting your shortcomings is the first step to self-improvement.

• technical

  • take more advantage of python packages
  • improve git commit etiquette
  • R syntax (as I mostly code in python, R is known to me but the syntax does not come as fluent; but I am taking more R projects to fix this)
  • better markdown skills ( this pages serves as an example )

• soft-skills

  • presenting progress to non-technical stakeholders
  • delegating tasks to people

Contact

e-mail: [email protected]
LinkedIn: Serban Radulescu