Carlos A.

I build and maintain data systems that collect information from websites, APIs and public records, then turn it into structured data your team can use. Whether you need to track property listings, combine supplier catalogs or keep public-record data up to date, I can help with the full workflow: source assessment, collection, matching, quality checks and delivery to your database or application. My strongest experience is in real estate, public records, retail catalogs and multi-supplier data. I work with the business rules behind the records: identifying the same property across sources, matching products from different suppliers, tracking changes and handling incomplete or conflicting information. Examples of my work include: • Processing 180,000+ property listings daily across nine Swiss real estate sources. • Monitoring 150,000+ retail products every 2–5 hours. • Experience working across 250+ county and public-record sources. • Classifying 50,000+ companies using LLM-assisted enrichment. For a powersports retailer, I built a multi-supplier data platform combining scraping and APIs, product normalization, cross-supplier matching, change monitoring, an authenticated operations dashboard and reviewable catalog exports. For a real estate intelligence project, I built an integrated console combining property-listing collection, agency matching, business-register enrichment and a valuation tool. Background jobs and run history let the client manage the data workflow through one interface. I can help you with: • Data collection — websites, APIs, dynamic pages, authenticated applications and difficult sources, with an approach suited to the required coverage and update frequency. • Data quality — normalization, deduplication, entity matching, validation and source traceability. • Recurring updates — scheduled and incremental collection, change detection, checkpoints, retries and alerts when something fails. • Data delivery — PostgreSQL, APIs, spreadsheets, catalog exports and integration with your existing platform. • Operational tools — dashboards for searching records, reviewing uncertain matches, managing runs and checking results. • Maintenance — diagnosing broken scrapers, adapting to source changes and improving existing pipelines. I also use AI-assisted extraction, classification and enrichment when they improve the workflow. The output is checked so uncertain or incomplete records can be identified and reviewed. I work primarily with Python, PostgreSQL, APIs and Docker, using tools such as Playwright, Scrapy and httpx according to the source. I assess source-access limitations and operating costs before committing to coverage or refresh schedules. You receive the source code, configuration and documentation needed to operate and extend the system. Engagements can include deployment, handover and ongoing maintenance. Send me the sources, the data you need, how often it should update and where the results need to go. I can help turn those requirements into a practical scope.