Data scientist and data analyst with experience in Python, SQL, machine learning, data visualization, and ETL pipeline development. Has worked on network health analytics, exploratory data analysis, A/B testing, reportin…
Data scientist and data analyst with experience in Python, SQL, machine learning, data visualization, and ETL pipeline development. Has worked on network health analytics, exploratory data analysis, A/B testing, reporting, and predictive modeling across telecommunications, insurance, fundraising, and home services domains.
- Enhanced network health of ~3M devices by developing a scoring algorithm with >15 attributes using modem traffic and performance data to identify customers with severe Wi-Fi modem issues. Employed feature engineering, including data
analysis, correlation studies, and domain expertise, to create features (e.g. broken fan indicator).
− Designed and implemented ETL pipeline within Databricks, to process raw network data and calculate KPI from diverse sources. Optimized processing and code to handle big data sets, reduced processing time by 6x. Delegate tasks to teammates.
− Streamlined data merger between Shaw and Rogers: Automated Shaw's data ingestion using Python and REST API, collaborated with Shaw's data team to standardize KPI calculations and ensure a unified KPI calculation methodology on both sides.
- Apply machine learning models on ad-hoc requests.
Enhanced network health of ~3M devices by developing a scoring algorithm with >15 attributes using modem traffic and performance data to identify customers with severe Wi-Fi modem issues.
Employed feature engineering, including data analysis, correlation studies, and domain expertise, to create features such as broken fan indicator.
Utilized Python libraries and SQL for analysis and implementation.
Designed and implemented ETL pipeline within Databricks to process raw network data and calculate KPI from diverse sources.
Optimized processing and code to handle big data sets (>100M rows daily), reducing processing time by 6x.
Automated the pipeline with dynamic backfilling for days with missing data.
Streamlined data merger between Shaw and Rogers.
Automated Shaw's data ingestion using Python and SharePoint REST API.
Collaborated with Shaw's data team to standardize KPI calculations.
Broke down Scala and Python code of the KPI tables to ensure a unified KPI calculation methodology on both sides.
Applied machine learning techniques such as k-means, DBScan clustering, and anomaly detection to establish statistically validated thresholds for network KPIs.
Used Python library scikit-learn to identify when performance starts to degrade.
All tables are processed and deployed from non-production to production environment.
Won Ted Roger’s Customer 1st Award.
Decreased customers calling by 27% and in-home service truck requests by 28%.
Identified 79K devices with broken fans and 4K with ventilation issues.
Corrected 83K customers that were given incorrect tier speed.
Transformed vague questions into structured analyses by performing exploratory data analysis in Python.
Clearly defined metrics calculation and its importance for decision making.
Strategically launched features on defined user groups.
Conducted A/B tests and T tests to evaluate key metrics outcomes.
Ensured a positive impact before full deployment.
Visualized findings on Metabase using SQL.
Optimized data presentation.
Generated informative, interactive Tableau reports for top management.
Provided concise insights and drill down options.
Automated reports for actuary team, including loss triangle analysis and future loss visualization using SQL.
Improved risk assessment efficiency.
Assisted in data mart design by defining table specifications.
Created logical schemas.
Documented source-to-mapping for tables.
Conducted high-level exploratory data analysis and source comparison to assess data quality.
Ensured data integrity.
Utilized R-based predictive models for fundraising improvements.
Analyzed donor behavior across all Canadian disasters, including the 2019 BC wildfires.
Conducted data cleaning, feature research and engineering.
Explored diverse machine learning models such as decision tree, random forest, and logistic regression.
Optimized models via backward/forward selection for efficiency.
Assessed performance using F1-score, recall, or precision as per the use case.
نظام التوظيف التشغيلي المعتمد على الذكاء الاصطناعي أولاً، الذي يوحّد ويؤتمت دورة التوظيف بالكامل للمؤسسات الحديثة حول العالم.
د.جوب منطقة حرة ذ.م.م. 2026 © جميع الحقوق محفوظة
