Emil A.

Led the development of an agentic AI pipeline for automated stool image analysis, integrating image alignment, segmentation, detection, and classification. Designed a hybrid workflow combining YOLO for bounding box detection, SAM (Segment Anything Model) for segmentation, and advanced alignment methods (SuperGlue, DGC-Net, GLUNet) to enhance region-of-interest precision. Fine-tuned SAM with a custom COCO-annotated dataset and trained a U-Net CNN for disease-relevant feature extraction and classification. This end-to-end system supports early illness detection in pets through robust and reproducible image analysis workflows.

Developed an interactive Power BI dashboard integrating GIS data to analyse employee commuting distances to office locations. Automated geospatial data processing in Python to calculate travel distances and identify optimisation opportunities for workforce allocation. Delivered actionable insights that supported operational planning and improved decision-making efficiency. Collaborated closely with cross-functional teams to ensure data accuracy and usability for management reporting.

Developed customised Power BI dashboards and reporting tools for multiple branches of a leading Dutch network provider. Automated data integration and transformation processes to deliver near real-time operational insights across sales, ticketing, and service performance. Designed scalable reporting frameworks, enabling branch managers to track KPIs and improve network reliability. Enhanced decision-making efficiency by standardising reporting workflows across the organisation.

• At Tutorful, he collaborated with the Data Platform team to expand and optimize data models, enabling over 15 new use cases and ensuring seamless integration across systems.

• Using AWS and SQL, he enhanced data ingestion and processing capabilities within the data lake, data warehouse, and ML platform.

• He implemented data quality tools and processes, achieving a 30% improvement in data accuracy and reliability, while designing future-ready solutions that anticipated the evolving needs of the internal ML platform.

• Leveraging AWS Quicksight, he developed scalable and cost-efficient data infrastructure improvements, optimizing performance by 25% and enabling more effective data-driven decision-making across the organization.

• As a Senior Data Specialist at The Fruit People, he led three major projects that transformed data management, demand forecasting, and workforce optimization.

• First, he designed and implemented a data integration pipeline, extracting information from employee and client management systems, delivery routes, and sales platforms and centralizing it in a robust Azure database.

• Second, he developed a time series forecasting model to predict demand for each SKU and store, paired with a Streamlit desktop app for the sales team, improving procurement accuracy and sales planning.

• Lastly, he created a mathematical optimization algorithm for an HRMS, automating weekly task scheduling and enabling strategic day-off planning.

• This solution minimized planning time, mitigated overtime risks, and highlighted skills in demand, aiding targeted hiring and training decisions.

• These initiatives significantly enhanced operational efficiency, improved demand accuracy by 25–30%, and reduced workforce planning time by 50%.

• At Uptime, I ensured the integrity of ETL pipelines and conducted analytical experiments to address challenges across various domains.

• He developed multiple dashboards in Tableau to provide stakeholders with actionable insights.

• Leveraging machine learning, he built a classification model to categorise schemas and handpump subsidy categories and a regression model to predict required subsidies and potential revenue.

• To streamline access, he created a Streamlit app for data analysis and model interaction. Additionally, he contributed to a scientific paper, highlighting our findings and their implications.

• These efforts improved decision-making, enhanced predictive accuracy by 20–25%, and provided stakeholders with clear, data-driven insights to optimise subsidy allocation and revenue strategies.

• At Gosh Food, he developed end-to-end data pipelines to streamline the processing and analysis of key FMCG metrics.

• He calculated and monitored critical KPIs using a custom-built system, enabling management to track real-time performance trends.

• Leveraging time series modeling, I provided actionable insights into sales and operational patterns.

• Additionally, he established data governance regulations and created comprehensive data dictionaries, ensuring consistency, accuracy, and compliance in all data-related processes.

• These initiatives improved data accessibility, reduced reporting time by 30%, and enhanced decision-making across the organization.

• At Nordan Construction, he designed and implemented an ELT pipeline to integrate data from various sources, including web scraping and API integration, creating a centralized real-time tender database.

• Collaborating with sales and engineering teams, he developed a predictive model to identify high-priority tenders ('hot call' tenders) and integrated this information into an interactive dashboard.

• This allowed the sales team to act swiftly, significantly increasing the likelihood of securing tenders by being the first to respond.

• These initiatives enhanced operational efficiency, reduced manual data processing by 40%, and streamlined tender tracking and decision-making, directly improving business outcomes.

• At Virtual Concept, he spearheaded the development of a fully automated Enterprise Resource Planning (ERP) and Customer Relationship Management (CRM) system from scratch, leveraging APIs to integrate various platforms into a centralised database hosted on a VPS.

• The solution streamlined core business processes, automating invoicing, generating dynamic resource and customer reports, and implementing recommendation engines, among other features.

• This project improved operational efficiency by an estimated 30–40%, reduced manual workload by 50%, and enhanced customer relationship insights, enabling data-driven decision-making across the organisation.

• As a Technical Data Analyst at KONE, he analyzed over 60 datasets from the technical help desk to identify bottlenecks, leading to an estimated 20% reduction in response time.

• He developed an interactive dashboard that provided real-time insights into key support metrics, enabling the management team to make data-driven decisions and improving resolution rates by approximately 15%.

• Through statistical analyses, he uncovered trends that optimized workflows, contributing to a measurable increase in overall service performance and customer satisfaction.

• Identificatie van strategische risico's en bijdragen aan risicobeperkende strategieën.
• Vaststellen van de relevante statistieken en prestatie-indicatoren om vooruitgang te meten.
• Advies over nieuwe engineering-praktijken en technologieën.
• Verbeteren van de Agile-praktijken.
• Opbouwen van relaties en allianties met het oog op sterke samenwerkingen.

• Vertaling van feedbackvereisten in een testbare hypothese, opstellen van proefschriften en het creëren prototypes voor pipelines die data engineers kunnen ontwikkelen en optimaliseren voor toepassing in machine learning door datawetenschappers.

• Verzamelen en analyseren van historische en huidige data met behulp van relevante computersoftware.
• Manieren identificeren om de operaties te verbeteren en de kosten te beperken gebaseerd op data-analyse.
• Een team van datawetenschappers leiden.
• Dataprojecten plannen en beheren.
• Tools ontwikkelen voor strategische besluitvorming.
• Onafhankelijk onderzoek uitvoeren met probleemdefinitie, onderzoeksplan, experimenteel ontwerp, ontwikkeling van algoritmes, code-implementatie en benchmarking.
• Implementatie van innovatieve voorspellende analyse-oplossingen op het gebied van financiën, logistiek en verkoop.
• Analyse van onderzoeksverslagen.

• Responsibilities: Including collecting past and present data, analyzing collected data, using relevant computer software, finding ways to improve operations and reduce costs, managing a team of data scientists, planning data projects, and providing the tools needed to make strategic decisions.

• Recruitment and Education management of Junior Data scientist Conduct independent research which includes problem definition, research plan, experimental design, algorithm development, code implementation, and robust comparison to available benchmarks

• Lead a portfolio of Data scientists in various projects

• Introducing cutting-edge predictive analytics solutions in multiple business areas including finance, logistics, and sales.

• Research papers analysis.