Data Science, Analytics, Visualizations, Machine Learning
My name is Ivan and I am based in the Washington metropolitan area.
I am a data scientist with a passion for storytelling. I enjoy working on data-driven applications, websites, products and projects that encompass data modeling, mining, simulations, visualization, and dashboard design.
I provide value to startups, organizations and teams by:
I earned my Bachelor of Science Degree in Computational and Data Sciences from George Mason University.
I come from Eastern Europe, Bulgaria and in my free time I enjoy singing, playing instruments, writing and recording songs.
I am available for new projects and work, contact me at ivan@ivanstoykov.com.
Interesting links, tools, and resources on big data
Insights, tips and articles on working with big data
Great tools and services for experimenting and interact with data
Here are some of the projects I've worked on and could share
I designed a real-time litter recognition system using the YOLOv5 deep neural network that recognizes plastic, metal, and glass. The models were trained using open-source images and achieved recognition of upwards of 70% for plastic, metal, and glass. This model was packaged into a Rest API and submitted to the 2021 Opportunity Project Competition hosted by the CENSUS Bureau.
My team and I extracted, transformed, and unified data from the top 5 global plastics pollution datasets - EarthChallenge 2020, Litterati, Cleanswell, TIDES and Marine Debris Tracker. We designed an interactive dashboard showing the statistical difference in plastic pollution before and after enacted state legislation. I also created a presentation and presented findings in a live-streamed event alongside Harvard and Villanova students.
Read article written about our project by Dr. White from George Mason University
I designed a custom 3D printable syringe pump system for precise delivery or extraction of liquids using Fusion 360. The pump featured a manual pusher release button, and was designed to fit disposable syringes of any size between 3 and 60 ml without additional, swappable attachments. I programmed and soldered the SMD stepper-motor driver to control it.
I enjoy solving problems, collaborating with others and helping startups and organizations. If you’re working with data and need a data-driven solution or if are interested in hiring me, get in touch.