Data Quality Intern - SQL & Validation

Start with the bad rows

In March a duplicate employer record sent the same rejection email to one candidate four times. Nobody spotted it for nine days. This two month internship exists so that problems like that get caught by a test instead of by a complaint. It runs remote within India, with one short check in each morning.

Your remit

You will profile the core tables, candidates, employers, postings and applications, and write up exactly where the data breaks: missing city values, phone numbers sitting in the email column, applications pointing at postings that no longer exist. Then you turn those findings into automated checks, wire them into the nightly run, and build a simple scorecard so each team can see the health of the tables they own. Fixing the worst offenders at source is part of the job too, not just reporting them.

Who does well here

People who are slightly bothered by inconsistency. You will need SQL, a systematic mind and the confidence to tell a team their data is wrong. There is no experience requirement, final year students are encouraged to apply, and the work is genuinely visible because everything downstream depends on it.