Zafran W

Python exertisation

Adobe • Freiberufler

Jun 2025 - Present • 11 mos

The start of this period was defined by high-stakes problem solving for international clients requiring immediate digital transformation. My primary objective was to take legacy, manual processes and replace them with high-performance Python architectures. One of my standout projects involved a real estate firm struggling with manual data entry from over 50 different property portals. I engineered a multi-threaded web scraping ecosystem using Scrapy and Selenium. Unlike standard scripts that crash when a website layout changes, I implemented a dynamic element detection logic that allowed the scraper to adapt to UI updates automatically. This reduced the client’s data collection time from 40 man-hours per week to just 15 minutes of automated execution. To ensure the data was usable, I integrated a Pandas pipeline that cleaned, de-duplicated, and formatted the data into a custom SQL database. This wasn't just a script; it was a business-saving asset. Midway through the month, I was commissioned to solve a complex API integration challenge for an e-commerce startup. They needed to sync inventory across three different platforms (Shopify, Amazon, and eBay) while managing fluctuating currency rates. I built a custom middleware using FastAPI that acted as a central nervous system for their business. By utilizing asynchronous programming (asyncio), I ensured the system could handle thousands of requests per second without lag. My ability to deliver this within a 10-day sprint proved my "best-in-class" technical agility.