Building data-driven systems for smarter decisions across SEO, Ads, Supply Chain and Operations using Python, SQL and Power BI.
I build AI-driven decision systems that help e-commerce businesses improve growth, inventory visibility and operational performance.
I focus on transforming raw business data into scalable and actionable decision-making systems using Python, SQL, Power BI, Google Ads, GA4 and ERP integrations.
Specialized intelligence systems for e-commerce growth and operations.
Predictive and optimization systems designed for smarter business decisions.
Stock visibility, replenishment risk and operational analytics systems.
Google Ads, SEO and GA4 analytics systems focused on growth and performance optimization.
SQL → Python → Power BI workflows and automation systems.
Business-focused dashboards supporting strategic and operational decisions.
AI-driven analytics and decision systems for e-commerce growth and operational visibility.
Executive Summary: A marketing decision intelligence system designed to transform fragmented marketing data into structured, data-driven business decisions.
Problem: Disconnected datasets meant manual, reactive, and inefficient decision-making.
Solution: A Python pipeline handling data processing, feature engineering, and decision logic to generate actionable recommendations for marketing optimization.
Impact: Reduced manual analysis workload by over 40 hours per month, prioritizing high-value customer segments and campaigns over intuition.
Executive Summary: A machine learning pipeline shifting organic SEO from reactive analysis to a data-driven, predictive decision system managing 50,000+ URLs.
Problem: Analysts spent 40+ hours per month reviewing Search Console data without predictive visibility or ROI frameworks.
Solution: Processed 50k+ URLs, engineering 30+ features, computing ROI on predicted traffic value against implementation cost, and leveraging ML Random Forest to forecast clicks.
Impact: Increased organic traffic value by 35% through ROI prioritization and automated the equivalent of 40+ hours of manual analysis.
Executive Summary: A machine learning pipeline to transition Google Ads budget management from manual, reactive updates to automated, data-driven strategies.
Problem: As ad spend grew 65% YoY, manual budget decisions on multi-channel, holiday-spike sensitive data failed to systematically forecast scale scenarios.
Solution: Random Forest Regressions simulate 5 budget levels to generate explicit AI-scored recommendations directly via Ads API, backed by Anthropic context models.
Impact: Kept overall ROAS stable at ~14.6x while managing a 30% YoY revenue uptick and heavily streamlining budget redistribution decisions across 4 distinct ad-channels.
Power BI dashboards designed to support executive decision-making, performance optimization, and operational visibility.
Technologies and concepts I leverage daily.
Continuous learning in data analytics, business analysis, and supply chain.