Data Harvest & Collection Engineer
Engineer · Remote or hybrid · full-time or long-term contract · AI-copilot proficiency required
The backbone of our data pipeline — building and operating the collection infrastructure that sources, deduplicates, and archives entities across the global longevity and health-technology landscape.
What you'll do
- Build and maintain web collection infrastructure: crawlers, parsers, and scrapers across structured and unstructured sources
- Handle the hard parts of the open web: bot-detection bypass, browser fingerprinting, proxy rotation, rate-limit management, and anti-bot countermeasures
- Source entities at scale — clinics, companies, labs, devices, vendors, scientists, physicians, and investors — from registries, databases, and the open web; deduplicate and normalize output
- Integrate structured data sources (corporate registries, regulatory databases, public filings) alongside open-web crawls
- Keep the archive clean: deduplication logic, source provenance tracking, confidence flags, and structured output schemas
- Own the collection stack end to end
What we look for
- Solid backend engineering background
- Real crawling and parsing experience: you've shipped scrapers that work against resistant targets, not just requests + BeautifulSoup on static pages
- Familiarity with anti-bot layers and practical techniques to work around them (fingerprinting, headless browsers, proxy infrastructure)
- Structured thinking about data quality — provenance, deduplication, confidence, schema discipline
- Comfortable working remotely and self-directed
Nice to have
- Experience with corporate/regulatory registry APIs or bulk data (company registries, EDGAR, ICIJ, OpenCorporates)
- Familiarity with identity resolution across heterogeneous sources
- Working proficiency with AI copilots as daily tools
AI-copilot proficiency (hard requirement). You use AI assistants as working instruments — to research, structure, draft, and verify — not as a question-and-answer chatbot. The screening questionnaire checks for exactly this.
Applying takes two steps — your details, then a short screening questionnaire with a few practical collection tasks — so we can see how you actually approach sourcing, parsing, and data quality.