Beschreibung

Competent Data Scientist with over six years of experience in Python programming and diverse machine learning methodologies. Proficient in developing and deploying various machine learning models, leveraging natural language processing (NLP), Large Language Models (LLMs), deep learning, statistical modeling, and traditional machine learning and optimization techniques to solve complex problems. Strong communicator and collaborator with a dedicated work ethic, delivering effective solutions. Passionate about leveraging AI for social good, driving innovation, and deeply committed to making a positive impact. Dedicated to continuous learning, and enjoys cooking, sports, and classical music.

Sprachen

Englisch
Verhandlungssicher
Deutsch
Konversationssicher
Hindi
Muttersprachlich oder zweisprachig

Arbeitsortpräferenzen

Vor Ort möglich

Hamburg (bis zu 50 km), Berlin (bis zu 50 km), Munich (bis zu 50 km), Düsseldorf (bis zu 50 km)

Freelance
Data Scientist
DIGITALAGENTUREN & IT-CONSULTING
Mai 2022 - Heute (4 Jahre und 1 Monat)
- Working as an AI for Good Engineer at Fruitpunch.ai, leading the development of data pipeline handling from preprocessing to zero shot prediction/labeling for a text-based environmental impact assessment using NLP and Large Language Models (LLMs).

- Collaborated with HAW Hamburg on satellite imagery analysis near the Ukraine-Russian border using QGIS, assessing vegetation health and organizing workshops for the faculty on GIS exploration fundamentals.

- Provided fraud detection expertise as a consultant for a BNPL Fintech startup, delivering reports and recommendations that improved the fraud detection rate by over 50%.
Python Machine learning NLP Large Language Modeling LLM ChatGPT EDA data pipeline Data analysis LCA QGIS Geospatial QGIS AWS Amazon Redshift Scikit-learn Hugging Face FastAPI Git CI/CD Docker Matplotlib Data visualization Dashboards Prototype Prompt Engineering
Sezzle EU
Data Scientist
E-COMMERCE
September 2021 - Mai 2022 (8 Monate)
- Developed and deployed end to end machine learning-based fraud detection models from scratch for card payment transactions, reporting to the Chief Product Officer. Utilized AWS (Redshift, S3, SageMaker) for data analysis & Modeling and FastAPI for containerization.

- Worked as the sole data scientist, collaborating with internal stakeholders throughout the modeling process. Validated data models with Pydantic and ensured REST API functionality through application containerization.
Python numpy Pandas Scikit-learn Data science SQL PostgreSQL AWS Amazon Redshift Docker Linux FastAPI API Jupyter EDA Data analysis Data visualization Dashboard
FREE NOW
Data Scientist
TRANSPORTWESEN
Mai 2019 - August 2021 (2 Jahre und 3 Monate)
Hamburg, Germany
- Developed an improved driver churn prediction model, resulting in a 25% reduction in driver churn rates.

- Aligned GMV forecasts with business goals, achieving a 20-60% reduction in forecasting error and optimizing revenue projections.

- Enhanced destination selection on smartphones by achieving a 38% improvement in Mean Reciprocal Rank (MRR) for personalized recommendations.
Python Airflow ETL Pandas Scikit-learn PostgreSQL SQL Apache Presto PySpark MLFlow Metabase Redash Dashboards Miro Git GitHub Atlassian JIRA CI/CD Continuous integration Docker Jupyter