Skip to main content

DataEngineer.fi | Portfolio

Iā€™m Mirlan, a Data Engineer & Data Analyst šŸ“Š with a strong background in Google Cloud Platform (GCP) services. I help businesses make data-driven decisions by building scalable data solutions, cloud data warehouses, designing efficient ELT pipelines, data architectures, and insightful BI solutions.

This is a page with real cases of some of my work. You can contact me using a form on the home page DataEngineer.fi

ELT pipeline: ROAS aggregate table (OLAP) and Visualizationā€‹

ROAS aggregate table (OLAP) and Visualization. ELT pipeline that includes:

  • API extract of costs from Meta Ads, Google Ads
  • API extract of Ad Revenue from Appodeal, AppLovin
  • in-app purchase revenue based on Firebase in-app event
  • Pipeline design in Dataform to get OLAP data connected to BI

ROAS metric ELT pipeline

Kimball Data architecture implementation for Mobile gamesā€‹

  1. ELT pipeline design in GCP (Dataform, schedulers, Cloud functions)
  2. Data integration (sources: API, S3 bucket, Firebase, BigQuery)
  3. Master Data table design
  4. Data Marts design
  5. OLAP cubes for BI instruments (Tableau, Metabase)
    • metrics: Retention, Churn, Level progress, Installs, DAU, sessions, traffic source, ROAS, etc.
    • cohorts: install dates, ad campaigns, countries, devices, depositors, etc.

Kimball for Mobile game

ELT pipeline: Installs table (master data modelling)ā€‹

Installs table (master data) for Mobile game with attribution. ELT pipeline based on sources:

  • RAW in-app event data to define Installs
  • User level Attribution data (fetched from Appsflyer by API or Singular)

Installs table

Metabase an open source BI instrument setup on GCPā€‹

Metabase docker container launched in Google Cloud Run and connected with PostgreSQL DB (needed for meta data) using Google Cloud SQL service.

Metabase

Automatic creation of In-app purchases in Appstoreconnect, through APIā€‹

I have created a Python service that handles a batch creation of in-app purchases in AppStoreConnect through API. Service is managed from Google Sheet (UI created using AppScript).

IAP creation in Appstore

Telegram channel parserā€‹

Telegram channel parser:

  • Scheduled Python service launched in docker in GCP and loading data to BigQuery

Telegram parser