Fetcher

Company overview

Fetcher (legally Tiplinks, Inc.) is a New York-based AI talent-sourcing platform, founded around 2014 and rebranded to Fetcher in 2019, that automates passive and active candidate sourcing plus personalized email outreach for corporate recruiting teams. Its AI scans a large candidate database (marketed at 500M+ profiles) to generate candidate batches that human sourcers and recruiters review before outreach, and it offers DEI/diversity search filters and pipeline diversity metrics. The company has raised roughly $40M, serves 1,000+ organizations, and holds SOC 2 Type 2 certification. Pricing runs self-serve from about $115/month up to mid-five-figure annual enterprise contracts, placing it in the mid tier.

Regulatory exposure

As an AI sourcing and outreach tool used by NYC and multinational employers, Fetcher sits adjacent to NYC Local Law 144 (its screening/ranking features could constitute an AEDT for deployers), the EU AI Act's high-risk employment provisions, and Illinois/Colorado AI-hiring rules. Despite this, Fetcher publishes no bias audit, no AEDT compliance guidance, no FRIA support, and no AI-specific technical or explainability documentation. Its diversity features (DEI filters and demographic pipeline metrics) also involve protected-characteristic processing, though the privacy policy states sensitive categories are excluded. Compliance obligations are pushed to customers: the terms make subscribers responsible for laws related to illegal discrimination.

Path to a higher score

Fetcher could raise its score by commissioning and publicly posting an independent NYC LL 144-style bias audit of its matching and screening AI (ideally two consecutive years), publishing a model/system card plus an explainability statement for how candidate matches are produced, and adding EU AI Act deployer guidance with a FRIA template. Restoring a first-party security/trust page with a subprocessor list and public DPA, standing up a status page and a vulnerability-disclosure channel, and documenting concrete in-product human-oversight controls (override, audit logs, per-jurisdiction toggles) would lift the technical-documentation, oversight, monitoring, and customer-documentation categories.