نبذة عني
Applied Economics and Data Science professional with experience in data science, econometrics, and business analytics. Skilled in Python, SQL, statistical modelling, and predictive analytics, with experience at ABB Robot…
Applied Economics and Data Science professional with experience in data science, econometrics, and business analytics. Skilled in Python, SQL, statistical modelling, and predictive analytics, with experience at ABB Robotics and in academic research on innovation and patents. Open to international opportunities.
الخبرة
Data Scientist
<p>Independently extracted, explored, and analysed approximately 14 million rows of SAP data using Python and Snowflake SQL to identify relevant cost drivers for robot configuration options.</p><p> Developed a cost estimation model using pairwise analysis and hedonic pricing regression to estimate prices for configuration options without directly available cost information, analysing 83 robots and over 1,100 configuration variants.</p><p> Validated model estimates with an RMSE equivalent to approximately 10–15% of the observed cost range.</p><p> Developed and deployed a Streamlit application enabling the team to retrieve estimated costs and descriptions for robot configuration variants, automating calculations and saving approximately four hours per variant.</p><p> Collaborated with teams in Sweden, Germany, and Spain and presented the analysis and solution to senior leadership.</p>
Marketing Data Analyst (Part-time)
<p>Analysed website search performance for Volcano Trails, an Iceland-based travel company, examining search queries, clicks, impressions, click-through rates, and average positions to identify high-performing opportunities.</p><p> Conducted country-level market analysis and identified the United States, Ireland, and Canada as priority markets, informing recommendations for geographically targeted advertising.</p><p> Contributed to data-driven campaign targeting strategies during the June–September peak season; overall sales increased by 27% year-on-year, with the identified priority markets among the major contributors.</p><p> Collaborated with the BI team to develop Looker Studio dashboards tracking revenue, advertising KPIs, marketing spend, budgets, and sales performance.</p>
Research Assistant – Applied Economics
<p>Contributed to Prof. Deborah Strumsky’s broader research project on climate adaptation innovation, investigating how inventor-team composition and knowledge diversity relate to invention novelty.</p><p> Suggested additional explanatory variables and regression specifications, collected inventor education-background information, and conducted logistic regression analysis using approximately 9,743 USPTO climate-adaptation patent observations.</p>
Data and Market Analyst Intern
<p>Conducted market and data analysis across defence and technology sectors, examining datasets related to explosive incidents, stockpiles, social media, and application performance.</p><p> Applied data cleaning, exploratory analysis, time-series analysis, regression, and predictive modelling to identify trends and support business and market research.</p><p> Researched potential competitors and analysed digital performance data, preparing reports and recommendations to support strategic and investor-related decision-making.</p>
المشاريع
Cost_variant calculation
<ul><li>Independently extracted, explored, and analysed approximately 14 million rows of SAP data using Python and Snowflake SQL to identify relevant cost drivers for robot configuration options. </li><li> Developed a cost estimation model using pairwise analysis and hedonic pricing regression to estimate prices for configuration options without directly available cost information, analysing 83 robots and over 1,100 configuration variants. </li><li> Validated model estimates with an RMSE equivalent to approximately 10–15% of the observed cost range. </li><li> Developed and deployed a Streamlit application enabling the team to retrieve estimated costs and descriptions for robot configuration variants, automating calculations and saving approximately four hours per variant. </li><li>Collaborated with teams in Sweden, Germany, and Spain and presented the analysis and solution to senior leadership.</li></ul>