The person behind the pipeline.

I'm Shiva, a software engineer working out of Kathmandu who spends most days convincing stubborn websites to hand over their data cleanly. Five-plus years in, my instinct is always the same: reach for the cheapest, most reliable extraction path first — an API if one exists, BeautifulSoup if the HTML is honest, and Selenium only when a site insists on rendering everything in JavaScript.

The extraction is only half the job. Raw scraped data arrives inconsistent — mismatched schemas, encoding artifacts, near-duplicate records — so I lean on regex, XPath, and Fuzzy Wuzzy matching to turn it into something a business or analytics team can actually use without a disclaimer.

Along the way I've also spent two years deep in QA and automation: taking API test coverage from 20% to 95%, and running mobile verification across Android and iOS on BrowserStack. That background shows up in how I build pipelines now — I assume edge cases exist until proven otherwise, and I write for the pipeline that has to survive contact with the real internet.