AI Governance
Last reviewed: July 6, 2026 · Broadlake Technologies LLC
HireProxy is a career-agent and interview-practice platform. It uses AI to help candidates represent their own professional background, analyze role fit, prepare for interviews, practice answers, and refine selected examples into recruiter-ready career evidence.
HireProxy maps its AI risk practices to the NIST AI Risk Management Framework 1.0 and the NIST AI 600-1 Generative AI Profile. NIST describes the AI RMF as a voluntary framework, and NIST has stated that AI RMF 1.0 is being revised. This page is a transparency summary, not a certification, audit report, or legal compliance claim.
The public documentation intentionally stays at the governance level. It does not expose raw AI instructions, internal evaluation cases, incident details, source code, database schema, tenant data, or provider-console settings.
What HireProxy Uses AI For
- Answering recruiter questions on candidate-owned career sites
- Comparing a job description to candidate-provided evidence
- Generating role-specific interview prep and study cards
- Transcribing spoken practice answers
- Providing spoken and written interview coaching
- Suggesting private career-story improvements for user review
What HireProxy Does Not Do
- HireProxy does not make hiring decisions.
- HireProxy does not rank, reject, or screen candidates for employers.
- HireProxy does not independently verify candidate credentials or work history.
- HireProxy does not infer identity, affect, stress, or personality from a user's face or voice.
- HireProxy does not publish raw Interview Mode answers unless the user chooses to refine or share them.
Current Controls
Grounded career data
HireProxy is designed around candidate-provided resumes, career stories, work history, job descriptions, saved prep, and interview transcripts. The AI is instructed to acknowledge missing information rather than invent facts.
Candidate control
Candidates control their career-site data, publishing status, interview prep, private practice attempts, exports, and account deletion. Raw interview practice answers are private by default.
Human decision boundary
HireProxy is not an employer screening, ranking, rejection, or automated employment decision system. Visitors should independently confirm information before making hiring decisions.
Voice privacy boundary
Interview Mode processes audio for transcription and spoken coaching. Saved practice records store transcript, metrics, and feedback rather than a saved practice audio file.
Abuse and incident intake
Public candidate sites include a report-abuse path, and HireProxy maintains platform-level incident response documentation for AI quality, privacy, security, provider, and voice-practice incidents.
Provider and telemetry controls
HireProxy uses third-party AI and infrastructure providers with scoped data flows, application-level error monitoring, rate limits, and privacy-constrained diagnostics on non-sensitive marketing surfaces when enabled.
Public Governance Documents
Public AI RMF Documentation
NIST Alignment Posture
HireProxy uses NIST AI RMF as a practical risk-management structure: Govern, Map, Measure, and Manage. The highest-priority risks for this product are confabulation, data privacy, information integrity, information security, human-AI configuration, and third-party value chain risk.
The reasonable next step for this stage of the company is to keep the public governance package current while maintaining the detailed incident register, evaluation cases, and provider review checklist internally. That creates practical evidence appropriate to a small company without exposing implementation details.