About
Passionate and innovative Lead Data Scientist, adept at spearheading transformative initiatives within global organizations, envisioning and executing data strategies leveraging cutting-edge technologies.
With a proven track record of orchestrating and leading successful projects, I harness the power of AI, Big Data, Data Science & Machine Learning to drive innovation and achieve tangible and valuable results.
Experienced in leveraging Cloud Computing and Automation technologies, I navigate end-to-end projects with agility, adhering to the CRISP-DM process and Agile methodologies to deliver impactful solutions.
I'm naturally committed to drive innovation actively engaging in R&D, seeking novel solutions to stay at the forefront of new technologies, and inspiring my team to embrace learning, pushing boundaries and fostering a culture of creativity and innovation.
Tech stack:
- Programming languages: Python, R, SQL, Spark.
- Data Visualization tools: Microsoft Power BI, Google Looker Studio, Qlik, Tableau.
- Agile & Software development tools: Git, Azure DevOps, CI/CD, Scrum, Kanban.
- Cloud solutions: Microsoft Azure, Databricks (certified), AWS, GCP.
Lead Data Scientist
- Company: EssilorLuxottica
- City: Milan, Italy
- Degree: Master & Bachelor in Statistical Sciences
- Birthday: 11 May 1989
- Email Address: federico.decillia@gmail.com
- Phone: +39 340 9310562
Resume
Download CVEducation
Master in Statistical Sciences
2012 - 2014
University of Padua, Padua, Italy
Principal studies:
Descriptive & Inferential Statistics, Econometrics, Machine Learning, Big-Data Analysis, Data Mining,
Marketing, Market Analysis, Business Administration, Economics, ICT
Thesis: La relazione tra uomo e ambiente. Analisi della sua evoluzione nel tempo mediante un approccio multilivello.
LLP Erasmus
2012
University Pompeu Fabra, Faculty of Economics, Barcelona, Spain
Principal studies: Courses focused on statistics & Business Economics, attended in English & Spanish.
Bachelor in Statistical Sciences
2008 - 2011
University of Padua, Padua, Italy
Thesis: Grado d'indicizzazione dei salari nominali dei sindacati U.S.A e Target d'inflazione
ICT High School Diploma
2003 - 2008
I.T.I. Severi, Padua, Italy
Courses & Certifications
Microsoft Azure Data Scientist Associate (DP-100)
Microsoft (07/2022)
Microsoft Azure Fundamentals (AZ-900)
Microsoft (12/2021)
AI for everyone
Deeplearning.ai (02/2021)
Deep Learning Specialization
Deeplearning.ai (06/2020)
Machine Learning Specialization
University of Washington (07/2019)
Complete Git & Github Masterclass
O'Reilly (10/2018)
Data Science & ML with Python & Spark
O'Reilly (06/2018)
Python for Data Science
University of California San Diego (10/2017)
Intro & Intermediate Python for Data Science
Datacamp (07/2017)
Text Mining and Analytics
University of Illinois (05/2017)
Machine Learning for Data Science & Analytics
Columbia University (01/2017)
Tackling the challenges of Big Data
Massachussets Institute of Technology (09/2016)
SAS Programming, SAS Macro Language, SAS SQL
SAS Institute (09/2015)
Professional Experience
Lead Data Scientist - Retail Global
02/2021 - Present
EssilorLuxottica, Milan, Italy
- Leading the Global Retail team in leveraging AI and Data Science methodologies to innovate existing processes. Provided compelling use cases and successfully implemented end-to-end solutions, driving tangible results and enhancing operational efficiency.
Data Science Pro Freelancer
2020 - Present
- Verified Fiverr Pro freelancer in AI & Data Science, offering Consultancy services and concrete AI deliverables. Completed end-to-end Data Science projects for 50+ Fiverr clients, maintaining an impressive average top rating of 5/5 demonstrating high-quality deliverables and client satisfaction.
Senior Data Scientist
09/2017 - 02/2021
Nielsen, Milan, Italy
- Lead many international projects to implement end-to-end Machine Learning (Classification, Regression, Clustering, Time Series) in multiple countries
- Direct exposure to management & clients presenting business insights & analyses
- Chair of Italian Diversity & Inclusion Team
Data Scientist Consultant
04/2016 - 09/2017
Bip (@Sky Italy), Milan, Italy
- Member of the “Big-Data & Advanced Analytics” community.
- Developed and implemented a NLP Churn-Prediction project & Text-Mining analyses, used to improve Retention / Upselling / Downgrade Marketing strategies / campaigns & better profile the Customer Base
SAS Masterclass & Data Scientist Consultant
06/2015 - 04/2016
SAS Institute & Value LabMilan, Italy
- SAS Masterclass on SAS Base, SAS Programming 1 & 2, SAS Data Lab for Hadoop, SAS Visual Analytics, SAS Visual Statistics, SAS Enterprise Guide, SAS Management Console, SQL for SAS
- Data Scientist Consultant for Value Lab (SAS client): Implementation & administration of ETL & Data Analysis processes, identifying targets of interest & showing results on BI platforms using various SAS Data Mining & Visualization tools.
Data Mining Junior Associate
02/2015 - 06/2015
Bocconi Alumni Association, Milan, Italy
- Data Mining & Data Analysis using SAP-CRM, STATA Software, Microsoft Office Suite.
Tutor Buddy for Exchange students
2012 - 2014
University of Padua, Padua, Italy
Skills
Programming Languages
BI & Data Visualization
Other Analytics Software
Cloud/Distributed computing & DevOps
Operating Systems
Languages
Portfolio
Forecast + Anomaly Detection of Global Retail Stores sales
Deployed a scalable end-to-end pipeline for daily sales forecasting across 15.000+ Retail Stores. Simultaneously, implemented anomaly detection of past sales to detect unexpected spikes and emerging trends.
- Tech: Python (Polars, Functime), Git, Microsoft Power BI, Databricks.
- AI: Blazingly-fast functime model deployed in Databricks (tens of thousands of forecasts in seconds), Anomaly Detection model combining 6 different alerts into a Main Alert to identify really critical anomalies.
- Achievements: Successfully integrated the highly accurate Daily Forecast + Anomaly Detection model into Microsoft Power BI, deployed Databricks cloud model.
- Team: Lead a team of 4.
Store Segmentation Web-App
Ideated and developed a Python Web-App from scratch for the Global Retail Business team (100+ people). The app streamlines annual Store Segmentation activities, replacing a manual, Excel-based process with efficient Machine Learning Clustering.
- Tech: Python (Streamlit, Scikit-Learn), Git, Docker.
- AI: Mutliple ML Clustering algorithms (KMeans, Spectral Clustering, Agglomerative Clustering, DBSCAN) developed and made accessible to end users.
- Web-App: Custom-designed using Streamlit, offering full flexibility and customization for the Global Retail team.
- Achievements: Transitioned from an outdated Excel process to an interactive web-app. Enhanced efficiency by incorporating Machine Learning Recommendation, resulting in time savings for Retail Merchandisers.
- Team: Lead a team of 3.
Anomaly Detection for 1000+ Panel Performance KPIs
Deployed a scalable end-to-end pipeline that provides a sorted list of most relevant anomalies among 1000+ KPIs on a daily basis.
- Tech: Python, Git, Spotfire.
- AI: Anomaly detection model using Facebook Prophet.
- BI: Developed an interactive BI Dashboard in Spotfire.
- Achievements: Significantly reduced operational team effort by pinpointing only critical anomalies.
- Team: Lead a team of 3.
Forecast of Covid Closures in 40+ International countries
Developed & Deployed a scalable end-to-end pipeline that forecast covid closures at country level, recommending when to stop operations in a particular country.
- Tech: Python, Git, Spotfire, Google BigQuery, Google Data Studio, Docker.
- AI: Forecast Covid closures in each country, following spikes noticed in N° of diseases and deaths.
- BI: Interactive BI Dashboard developed in Google Looker Studio.
- Achievements: Daily forecast of covid closures supporting operations to take immediate actions in uncertain times, saving cost and avoiding spread of diseases.
- Team: Lead a team of 3.
Forecast of Viewing/Listening Behaviors of TV/Radio viewers/listeners.
Developed & Deployed a scalable AutoML pipeline that predicts user behavior (N° of minutes of video content watched) across four countries (with scalability to 20+ countries).
- Tech: Python, Git, Spark.
- AI: Regression (Linear, Lasso, Ridge, Huber, Elastic Net), Ensemble Models (Random Forest, Extra Randomized trees, AdaBoost, Gradient Boosting).
- Achievements: AutoML pipeline built from scratch, Individual forecast of viewing behaviors, enabled personalized content recommendations.
- Team: Lead a team of 6.
Churn Prediction of Panelists in international Media Panels
Deployed a scalable Churn Prediction model that assesses the likelihood of churn for each panelist in the upcoming months.
- Tech: Python, Git, Spark.
- AI: Logistic, Ensemble (XGBoost, Random Forest, Extra Randomized trees, AdaBoost, Gradient Boosting, Voting), Deep Learning (Multi-Layer-Perceptron).
- Achievements: Scalable deployment, enabled targeted retention efforts, cross-functional teams collaboration.
- Team: Lead a team of 4.
NLP Churn Prediction Model
Enhanced existing Churn Prediction model by integrating customer care call transcripts. Fine tuned the model leveraging NLP for better accuracy and profitability of Prevention, Retention, Upselling & Downgrade Marketing strategies.
- Tech: Python, KNime.
- AI: Classification algorithms (Logistic, Random Forest, XGBoost, SVM), NLP, Text-Mining.
- Achievements: Integrated call transcripts data, Improved Retention Rates, Promo Optimization, Cost Savings.
- Team: Lead a team of 2.
Contact
Location:
Milan, Italy
Email:
federico.decillia@gmail.com
Call:
+39 340 9310562