Alexandre Kitanidis

Maintaining and development of the company's data platform, working in a team of several data engineers with an agile workflow:

- Implementation, orchestration and improvement of data pipelines

- Implementation of a near real-time data pipeline

- Connection of data pipelines to external APIs

- Work on cost saving solutions for S3 and Snowflake

- Migration of Airflow hosting solution to Kubernetes

- Support to the BI and data analysis teams for new data ingestion in Snowflake and troubleshooting of data related issues

Work on the full implementation of a data architecture for a new factory, from the definition of all the tables needed to be created in our cloud warehouse (on Snowflake) with an ELT tool (Matillion) to the different visualization tools that had to be created with PowerBI from these tables. Deployment of connected sensors in our factories, optimization of their data transmission to a SQL database connected to our cloud solution. Improvement of the tools to visualize and get alerts from the sensors data both with PowerBI and Google Apps Scripts. Improvement of an image analysis tool for larvae counting in an existing factory and development of a graphical user interface for the workers to use the tool. Development of several image and video analysis tools for different R&D projects. Innovafeed is a biotech company that produces a new source of protein from insect rearing for animal feed, more specifically for aquaculture.

Project of identification of genetic markers from tumor classification on gene expression data :

- Reduction of dimensionality through PCA applied to genes expression

- Representation of the genetic data as heatmaps

- Binary classification (with or without tumors) with a CNN

- Work on the interpretability of the classification with gradCAM algorithm to identify genetic markers