Imad L.

I build data systems that run in production ... LLM extraction pipelines, automated ETL, and scraping infrastructure that doesn't break when the source changes. $30K+ earned on Upwork. 100% Job Success. Top Rated. 1,400+ hours delivered. What I've actually shipped: — An AI document extraction system for a construction asset platform: Gemini-powered pipeline that reads unstructured operation manuals and drawings, extracts structured asset data, and loads it into the client's platform. Deployed on Google Cloud Run with event-driven triggers. — A cricket analytics platform (Decidr Sports): real-time data extraction from live sources, processing pipeline, and decision-intelligence dashboards. 400+ hours, client endorsed for being "solution oriented" and "accountable for outcomes." — LLM-based data normalization for a US insurance brokerage: messy carrier data in, clean standardized records out ... replacing hours of daily manual work. — Dozens of scraping and automation builds: Scrapy, Playwright, Selenium, anti-bot handling, deployed with Docker on AWS and GCP, scheduled and monitored so they keep running after I hand them off. How I work: I ask the right questions up front, propose the simplest architecture that solves the problem, and deliver something documented and maintainable ... not a script that dies in a month. Clients keep me on retainer because things I build stay running. Stack: Python, LLM APIs (Claude, Gemini, GPT), Scrapy, Playwright, PostgreSQL, MongoDB, AWS (Lambda, S3, ECS), Google Cloud Run, Docker, Make, n8n. If you have messy data, a manual process eating your team's time, or a source you need extracted reliably ... send me the details and I'll tell you exactly how I'd approach it.