Mercor
Mercor builds custom AI agents for enterprises by mapping workflows, ingesting company knowledge, and deploying production-grade automation in 4–6 weeks. The platform captures organizational context through AI-moderated interviews, converts it into agent behavior specs with quality guardrails, and iterates on failures through automated and human-driven feedback loops. Suited for recruiting, finance, customer service, engineering, and business operations, Mercor's agents learn continuously and operate across existing tools like Slack, spreadsheets, and internal systems.
Mercor builds custom AI agents for enterprises by mapping workflows, ingesting company knowledge, and deploying production-grade automation in 4–6 weeks. The platform captures organizational context through AI-moderated interviews and internal data ingestion, then converts that context into programmatically defined agent behavior specs with built-in quality guardrails and observability. Unlike generic AI models, Mercor agents are onboarded on a company's specific tools, workflows, and standards—operating across Slack, spreadsheets, internal systems, and existing infrastructure. The system includes automated and human-driven feedback loops that continuously improve agent accuracy by catching failures and feeding corrections back into the system. Suited for recruiting, finance, customer service, engineering, and business operations, Mercor's approach eliminates silent failures through clear error surfacing and safe deployment guardrails. The platform is backed by world-class AI research and engineering talent, with demonstrated results including 90% end-to-end support resolution, 73% reduction in close time, and $1.4M in savings with ROI within 2 weeks for early customers.
The policy explicitly states that user information including interview recordings, transcriptions, responses, and profil
No training opt-out mechanism was identified in the legal documents.
High risk across multiple dimensions. Only use with public, non-sensitive data.
The policy explicitly states that user information including interview recordings, transcriptions, responses, and profile data are used 'To evaluate candidate interviews using AI and third-party services' and 'To improve our platform functionality.' There is no clear restriction on using this data to train AI models, and the broad language about AI evaluation and service improvement suggests training is likely occurring.
“To evaluate candidate interviews using AI and third-party services. To generate professional photos from interview images. To improve our platform functionality.”
The terms grant Mercor broad rights to user content without explicit limitations. Users must conduct all negotiations through the platform and cannot work with introduced contacts for 2 years, indicating extensive control. The privacy policy allows sharing of 'profile, resume, and interview data' with employers and service providers without clear scope limitations on how this data can be used.
“Your profile, resume, and interview data may be shared with companies using our platform. With Service Providers: We use third-party services for AI evaluation, image processing, and platform functionality.”
Data is extensively shared with third parties. The policy explicitly states sharing with 'Employers,' 'Service Providers' for 'AI evaluation, image processing,' background check providers, and professional networking sites. The scope of third-party access is broad and not limited, with minimal restrictions on how third parties use the data.
“With Employers: Your profile, resume, and interview data may be shared with companies using our platform. With Service Providers: We use third-party services for AI evaluation, image processing, and platform functionality.”
The policy states data is retained 'for as long as necessary' with vague criteria including 'improve our services and develop new features' and 'legitimate business purposes.' While deletion is mentioned, the conditions are unclear and retention periods are indefinite. Users can request erasure but it's 'subject to legal requirements,' creating ambiguity.
“We retain your personal data for as long as necessary to provide our services to you, comply with legal obligations, resolve disputes and enforce agreements, improve our services and develop new features.”
The terms do not explicitly address IP ownership of user-generated content or AI outputs. The non-circumvention clause and platform control over salary negotiations suggest Mercor retains significant control over user relationships and data. The absence of clear IP ownership language favors the platform provider.
“Worker acknowledges that Mercor has a proprietary interest in the relationships it has with its clients, partners, and contacts. Compensation terms offered through Mercor are final and binding.”
The policy mentions GDPR compliance and legal basis for processing under GDPR, but does not explicitly claim SOC2, HIPAA, or ISO certifications. While GDPR compliance is referenced, there is no evidence of third-party certification or audit verification of security and privacy controls.
“Under the General Data Protection Regulation (GDPR), we process your personal data based on the following legal grounds: Consent, Contract Performance, Legitimate Interest, Legal Obligation.”
Track and measure your AI tools with Olakai's observability platform.
Explore Olakai