Leo Soukko Email & Phone Number

My job is to take part in building new data platform and create commercial (B2B, B2C) insights. The main tasks include:• Building data pipelines from source to iron and from iron to bronze (Python)• Building data pipelines from bronze to silver and from silver to gold (SQL, DBT)• Developing applications to utilize data as the basis for decision making (SQL, PowerBI)• Helping colleagues with technical issues related to dbt, GitHub & SQL• Conducting analysesMy main tools:• SQL• Databricks• DBT• Python• GitHub• Power BI• Slack• Teams• Jira• Confluence

My job was to take part in building new data platform and create commercial (B2B, B2C) insights. The main tasks include:• Building data pipelines from source to iron and from iron to bronze (Python)• Building data pipelines from bronze to silver and from silver to gold (SQL, DBT)• Developing applications to utilize data as the basis for decision making (SQL, PowerBI)• Conducting analysesMy main tools:• SQL• Databricks• DBT• Python• GitHub• Power BI• Slack• Teams• Jira• Confluence

My job was to create consumer insights and take part in building new data platform. The main tasks include:• Building data pipelines from bronze to silver and from silver to gold (SQL, DBT)• Developing applications to utilize consumer data as the basis for decision making. (SQL, PowerBI)• Conducting consumer analyses (who are our heavy users, how do they compare to market heavy users, what kind of consumers are "churning", what kind of consumers are most unsatisfied and why etc.)My main tools:• SQL• Databricks• DBT• GitHub• Power BI• Slack• Teams• Jira• Confluence

Helsinki, Southern Finland, Finland

My job is to improve data driven decision making in the VR passenger traffic division. The main tasks include:• Developing applications to utilize data as the basis for decision making. (SQL, PowerBI)• Conducting data related ad hoc tasks and analyses• Improving BI processes, for example to reduce time used for application upkeep/maintenance• Improving data quality which leads to more precise decision makingMy main tools:• SQL• Snowflake• GitHub• Power BI• Slack• TeamsOther tools:• Excel

Espoo, Uusimaa, Finland

My job was to create Power BI reports based on predefined specifications for medical personnel using SQL and PySpark in Azure Databricks. Typical process was to:• Take a task from Jira backlog and read the documentation for the wanted report• Query and transform the data using SQL and PySpark and write new tables• Load the data to Power BI and visualize it according to the specificationI also created some internal reports, for example: • A script that scraped Power BI data refresh errors and visualized them in a Power BI dashboard• A script that parsed the browsers used by our clients and visualized them in Power BI• A script that read log files to check what kind of access privileges users had in reportsI also took part in a project where we tried to predict the severity of a certain disease for patients. We used supervised learning algorithms for that.My main tools:• PySpark• PowerBI• SQL• Azure Databricks• Jira• Confluence• Slack

Helsinki Area, Finland

I was part of a data analytics team working on projects and sales cases considering optimization and predictive maintenance.Sales cases I participated in:• Predictive maintenance for industrial client. Business problem was to predict from sensor data when the client's industrial machines would start to choke on dirt formed during their manufacturing process, so the client could clean those machines before the entire manufacturing line had to be shut down. We tried to analyze which sensors might indicate some changes before the shutdowns using data visualization, statistics and supervised machine learning. I also clustered the data to see if there were different clusters few weeks/days before and after the shutdowns.• Optimization of machine maintenance schedule. We created Demo for client to show our maintenance schedule optimization using dummy data similar to the clients supposed data.Projects I participated in:• Creating new graph based data model for a client. We created a small graph database for a client, so they could more easily see the connections between different data objects.• Analyzing high vibration effects on employees. My job was to create an ETL script that would read binary files with 5000Hz sensor data from Cloud Object Storage, down sample and write them into a SQL Database. The script was written in Python and there were 18 - 36 million rows per file for about 1000 files.Main tools:• Python• Jupyter Notebook• Watson Studio• IBM Cloud Object StorageOther tools:• SQL (DB2)• Linux• git• Cypher (Neo4j)

Master'S Degree, Business Analytics, 4.5

Master's Thesis: Neural network based binary and multi-class trading strategies using probability thresholds for trading actions on S&P.

Bachelor'S Degree, Financial Management, 3.73

Major: Financial Management Minor: Economics