Hi, I'm Umer Iqbal —
I turn raw data into data products.

Data Analyst & Analytics Engineer from Islamabad, Pakistan. I build extraction pipelines, dbt-modeled warehouses, automated data-quality tests, and dashboards people actually read — end to end.

SQLPythondbt Core DuckDBStreamlitPower BI TypeScriptGitHub Actions

View my GitHub See projects ↓

Featured Projects

Real data, real engineering — each project runs end-to-end with tests and CI.

DATA WAREHOUSE · FEATURED

PSX Analytics — Pakistan Stock Exchange Data Warehouse

Fault-tolerant Python extractor pulls 24,600+ daily price records for 20 PSX large-caps plus USD/PKR and SBP policy rates into DuckDB. A dbt layer ships facts & dimensions — including a Type-2 slowly-changing dimension and ASOF currency joins — guarded by 41 automated tests and freshness SLAs. A five-tab Streamlit dashboard serves the insights, refreshed daily by GitHub Actions.

📦 24.6k rows🧱 8 dbt models ✅ 41 tests🕰 SCD2 ⏱ daily CI

Stack: Python · yfinance · DuckDB · dbt Core · Streamlit · Plotly · GitHub Actions · Parquet

DATA PRODUCT · WEB-SCALE EXTRACTION

GigCraft — Fiverr Market-Research Studio

Types a niche, crawls the public Fiverr market, computes real keyword & market intelligence from the crawled sample, and drafts complete, compliance-checked gig packages with an AI layer — three architectures (Express web app, FastAPI growth system, standalone MCP server) in one monorepo.

🌐 web-scale crawling📊 keyword analytics 🤖 AI drafting

Stack: TypeScript · Express · Python FastAPI · SQLite · Gemini API

MARKET ANALYTICS · WEEKLY PIPELINE

Data Job Market Radar

Four public job-board APIs scraped every Monday → an append-snapshot DuckDB warehouse → dbt marts that quantify skill demand, salary bands, and posting lifecycles → a five-tab dashboard. Week-1 result: SQL is in 63% of postings, Python 53%, and GenAI/LLM skills already 22%.

📡 4 live sources🧠 51-skill taxonomy ✅ 25 tests⏱ weekly CI

Stack: Python · DuckDB · dbt Core · Streamlit · GitHub Actions

SPORTS ANALYTICS · STORYTELLING

PSL Analytics — every ball of the Pakistan Super League

357 matches · 83,799 deliveries (2016–2026) flattened from Cricsheet into a tested dbt warehouse: career leaderboards (Babar 4,380 runs), venue character maps, run-worm match replay, and an explainable empirical win-probability model learned from every chase in league history.

🏏 83.8k deliveries🎯 win-prob model ✅ 35 tests🔍 reconciliation test

Stack: Python · DuckDB · dbt Core · Streamlit · Plotly · GitHub Actions

Skills

What I use day to day.

SQLquerying · modeling

Pythonpandas · pipelines

dbt Coremarts · tests · snapshots

DuckDBlocal warehouse

Power BIDAX · reporting

Streamlitdata apps

Excelanalysis · modeling

Git / Actionsversioning · CI/CD

TypeScriptdata tooling

Storytellinginsights → decisions

About me

The short version.

I'm a data analyst & analytics engineer who likes owning the whole pipeline: extract → load → model → test → serve. My flagship project is a full data warehouse for the Pakistan Stock Exchange — live extraction, dbt models with SCD2 history, 41 automated data tests, and a dashboard. I also build data-powered tools (see GigCraft).

Right now I'm sharpening dimensional modeling, Power BI DAX, and semantic layers — and looking for Data Analyst / Analytics Engineer opportunities.

📍 LocationIslamabad, Pakistan

💼 FocusDA / Analytics Engineering

🧰 CoreSQL · Python · dbt

🌐 GitHubBilalkhank10

🤝 Open toFull-time · remote · contract

Hi, I'm Umer Iqbal —I turn raw data into data products.