Skip to main content
Purity KavwelePK
Open to opportunities

Purity Kavwele

@puritykavwele

I’m a data science and IT professional driving automation and predictive analytics.

Germany
Message

What I'm looking for

I’m looking to apply data science and MLflow/ETL automation to build predictive, research-ready models and analytics. I want a collaborative team that values continuous learning, scalable Azure/cloud delivery, and measurable impact on decision-making.

I’m a versatile and analytical IT professional focused on data-driven decision-making, process automation, and predictive analytics. In my current role as a Data Science Research Assistant, I developed advanced pharmacokinetic generative AI models for few- and zero-shot time-series forecasting in mixed-effects settings, including deep mixed Gaussian Processes and sparse GP variational methods.

I build ETL pipelines to extract and transform pharmacokinetic data from APIs (DrugBank, ADME, PK-DB) and then run statistical analysis to improve treatment-level insights. I leverage Comet and MLflow to streamline reproducible ML workflows, collaborated closely with domain experts on experiment design, and used model comparisons to boost predictive accuracy by 40% while quantifying model variance.

Earlier, I supported predictive modeling and intelligent automation as a Data Science Intern, developing data-driven insights with Python and SQL and contributing to scalable analytics solutions. I also bring strong IT and cloud execution from my work as an Azure Solutions Architect/M365 Administrator—designing Azure environments, integrating MLflow for batch training, and delivering Microsoft 365 architecture with identity and security governance (MFA, Conditional Access, RBAC). I’m motivated by problem-solving, continuous learning, and turning big data into actionable insights for research and innovation.

Experience

Work history, roles, and key accomplishments

UP

Data Science Research Assistant

Universität Potsdam

Apr 2024 - Feb 2026 (1 year 10 months)

Developed generative AI models for few- and zero-shot pharmacokinetic time-series forecasting using deep mixed Gaussian Processes and sparse variational methods. Built ETL pipelines from DrugBank/ADME/PK-DB APIs and performed treatment-specific statistical analysis to support downstream modeling and evaluation.

IA

Azure Solutions & M365 Admin

Iwave Africa

Jan 2021 - Mar 2022 (1 year 2 months)

Designed and deployed scalable Azure Machine Learning compute environments to support customer data science workloads. Configured and executed batch training jobs in Azure Machine Learning with MLflow integration for model training tracking.

IK

IT & Data Intern

ICT Authority Kenya

Oct 2019 - May 2020 (7 months)

Supported the design and implementation of ICT systems and applications to improve digital service delivery in public sector institutions. Performed data analysis and visualization using Python, Excel, and Power BI, and assisted with ICT initiatives including system monitoring, maintenance, and user-facing portal support.

SC

IT Support Intern

Simba Corp

May 2018 - Aug 2018 (3 months)

Provided on-site IT support, installing and troubleshooting PC hardware/software and peripherals including Cisco IP phones, printers, and scanners. Managed support tickets in Spiceworks helpdesk, performed preventive maintenance, supported SAP ERP migration documentation, and handled Active Directory user lifecycle tasks.

Education

Degrees, certifications, and relevant coursework

University of Potsdam logoUP

University of Potsdam

Master of Science, Data Science

Completed an MSc in Data Science and automated pipelines at the University of Potsdam.

Kirinyaga University logoKU

Kirinyaga University

Bachelor of Science, Mathematics and Computer Science

Completed a BSc in Mathematics and Computer Science at Kirinyaga University.

Find your dream job

Sign up now and join over 250,000+ remote workers who receive personalized job alerts, curated job matches, and more for free!

Sign up
Himalayas profile for an example user named Frankie Sullivan