Data Engineer in the making. I build pipelines that turn messy sources into clean, trusted data. Automating security and compliance data | 2 yrs data automation at Miele | M.Sc. Data Science, TU Dortmund
📍 Dortmund, Germany
Most data problems start before the model or the dashboard: sources that change shape, manual exports, and nobody sure which number is right. That's the part I like to fix, with pipelines that collect, clean and load data so the people downstream can trust it, and an auditor can too.
I'm finishing my M.Sc. in Data Science at TU Dortmund, with the thesis in 2027. Before that I did a B.Sc. in Computer Science & Engineering at American International University–Bangladesh. I'm also learning German (A1.2, and yes, it's still a work in progress).
Next step: a full-time data engineering role in Germany from summer 2027.
- KPS AG · Working Student, Information Security (Sep 2026 – now) Data pipelines for an IT security team: collecting and processing external data sources, traceable for audits.
- Miele · Working Student, Gütersloh (Mar 2023 – Feb 2025) Python pipelines that moved development process documentation into a central database with versioning, plus a Web2Py process editor (deployed on IIS) that the teams used to maintain it. Figma prototypes and user feedback rounds for the editor's interface.
- BRAC Bank · Officer, System Support and Service Operations (Dec 2018 – Nov 2021) Kept core banking systems running 24/7: backups, archival and data validation during maintenance. Built an AMC tracker that alerted owners before maintenance contracts expired, and replaced a paper daily log with a shared digital one.
- Labaid Group · Executive, HIS/ERP Implementation (Jan 2017 – Dec 2018) Data mapping, cleaning and migration from legacy systems to a new hospital information and ERP system, plus dashboards and reports for operational and management users.
- PixelNet Technologies · Web Designer Intern (Oct – Dec 2016) Front-end work on responsive websites (HTML, CSS, JavaScript).
Scrapes live job postings, extracts the skills they ask for with an LLM, and scores a CV against them the way an ATS does: by keyword match. Turns out an ATS reads keywords, not skill levels.
Python FastAPI SQLite BeautifulSoup Selenium
Python SQL ETL Data Pipelines Database Design Data Migration Data Validation Data Quality Web Scraping FastAPI SQLite Git Web2py Figma
You'll likely find me with a book. Favourites include The Alchemist, The Compound Effect and Sei Somoy. More on my Goodreads.

