Flipping the Script on Market Data
Nearly all data regarding the job market (ghosting rates, average time-to-hire, salary transparency) is owned and reported by employers and ATS vendors.
I built the Research Module inside AutoApply to create the first massive, candidate-side dataset. As the bot applies to jobs, it passively records "signals"—such as deceptive "entry-level" jobs requiring 5 years of experience, or forms that force illogical constraints.
To ensure absolute privacy, the module uses SHA-256 hashing with a local salt to anonymize company names and URLs. It strips all PII (emails, phone numbers) before data is written to the local CSV. It is strictly opt-in, giving candidates a safe way to contribute to a publishable academic dataset that exposes how the modern hiring market actually functions.