Data Analyst with hands-on engineering experience
Python · SQL · PostgreSQL · Power BI
ETL Pipelines · API Integration · Commercial Analytics
January 2026–Present
I develop Python and SQL data workflows, integrate ERP data and translate sales, customer and inventory questions into reporting requirements. My work spans ETL, data quality, API integration and Power BI analysis.
Indoor Sales Consultant · Dr. Nutrition
May 2022–January 2026
Retail experience in sales targets, supplier negotiations and team training gives my technical work a practical commercial foundation.
I combine Python and SQL programming with data engineering, analytical modeling and business intelligence. I build pipelines with FastAPI, PostgreSQL, RabbitMQ, Dramatiq and Docker, using validation, incremental checkpoints and retry handling to make data flows dependable. My portfolio also explores machine learning, public catalogue analysis and education technology.
Based in Dubai, UAE. Open to data analytics, business intelligence and data engineering opportunities.
From system integration and reliable pipelines to analysis and decision support.
Integration engineering, business intelligence and evidence-led analytical case studies.
A multi-pipeline ERP integration project with SOAP extraction, canonical PostgreSQL models and branch-specific delivery. Checkpoints, idempotency, durable queues and retries support recoverable processing; body-assessment delivery remains contract-pending.
View project on GitHubCombined classification, anomaly detection and transparent rules across 558,211 public claims from 5,410 providers. On held-out providers, a review queue captured 81.2% of potential-fraud labels while selecting 16.8% of providers. Built for investigation support, not automated fraud verdicts.
View project on GitHubInteractive dashboards built from a source dataset of 500,000+ retail transaction lines to explore sales trends, customer behavior, and product performance.
View project on GitHubA SQL pipeline that transforms raw transactions into structured analytical models using Bronze–Silver–Gold architecture.
View project on GitHubA Python and Playwright pipeline with retries, resumable collection and Excel export. The committed snapshot contains 10,188 distinct product URLs; brand and category coverage is incomplete and documented.
View project on GitHubAn experimental comparison of classification models for retrospective customer inactivity. Future churn prediction requires time-based validation; financial scenarios are not measured savings.
View project on GitHubLooking for help with ETL pipelines, system integration, dashboards or commercial analysis? Let’s discuss the data and the decisions it needs to support.