Research Intern
I had to opportunity to be a data science pioneer at one of critical biggest products, project Oversee. My task was the automatic classification of vessels as either fishing or not, given telematics data, an essential task to combat illicit fishing.For that purpose I created a distributed system for the aggregation of positional updates and automatic feature extraction, a system that will likely encompass other modules besides fishing detection in the future.The project went very well, and surprisingly I accomplished a very high accuracy, I feel like I have made a good contribution to the project, and that fishing restrictions and enforcement can now be tighter. Technologies used: Apache Spark, Kafka, Avro, Scala, Python