WadeCV

AI Disclosure

Last updated: October 22, 2026

How WadeCV Uses Artificial Intelligence

WadeCV uses artificial intelligence to generate tailored CVs and cover letters. We believe in transparency about how AI is used in our service.

AI Model

  • Model family: Anthropic's Claude Haiku 5.5
  • Provider: Currently Anthropic's Claude API, but this may change over time and we may use other AI providers
  • Purpose: Generating tailored CV content and professional cover letters

What the AI Does

  • Analyzes your existing CV content and extracts key qualifications, experience, and skills.
  • Reads the target job description to understand role requirements.
  • Rewrites and restructures your CV content to better match the target role.
  • Generates a professional summary tailored to the specific job.
  • Rewords experience bullet points to highlight relevant achievements.
  • Writes professional cover letters based on your tailored CV and the job description.

What the AI Does NOT Do

  • Fabricate experience: The AI is instructed to only enhance presentation of your actual experience, not invent roles, companies, or qualifications you don't have.
  • Train on your data: Anthropic processes your data to generate your documents and does not use your prompts or responses to train its models (per Anthropic's commercial terms). It deletes them within 30 days, or keeps them up to 2 years if its safety systems flag a request.
  • Make hiring decisions: on the job-seeker side WadeCV is a content creation tool. On the recruiter side (below) the AI scores CVs as a decision aid, and people make every hiring decision.

WadeCV for Recruiters

Organisations can use WadeCV to screen candidates. There, the same model reads a CV from the employer's side:

  • Rubric drafting: from a job description, the AI proposes the criteria to score against, how each is measured and how much it weighs. A recruiter reviews and edits the rubric before any CV is screened.
  • CV screening: the AI extracts contact details, summaries of education and experience, years of experience, location, the value of each rubric criterion and free-form attributes, and gives a fit score from 0 to 100 with written strengths and gaps. A second, arithmetic score is computed from the extracted values and the organisation's weights and can be recomputed by a recruiter without the AI.
  • Routing and search: a CV uploaded without a role is assigned to the most suitable open role; screened candidates are searchable by meaning through vector embeddings of their summaries; a chat assistant can search and summarise an organisation's candidates.
  • What it is instructed not to do: infer age, gender, ethnicity, religion, disability, nationality or family status; this instruction is tested. It does not predict tenure or performance, does not analyse candidate videos, does not reject anyone, and does not learn from an organisation's decisions.
  • Human in the loop: every score can be overridden and every stage change is made by a named member of the organisation. Applicants are told on the application form that AI reads their CV. Organisations are responsible for the notices, audits and assessments required where they hire.

Important Disclaimers

  • AI-generated content may contain errors, misinterpretations, or inaccuracies.
  • You are responsible for reviewing all generated content before using it in job applications.
  • AI may rephrase your experience in ways that don't accurately reflect your actual role, so always verify.
  • Generated cover letters are a starting point and should be personalized with specific details about the company.
  • AI output quality depends on the quality and completeness of the input you provide.

Your Control

You always have full control over the final output:

  • An inline editor lets you modify every section of the generated CV before downloading.
  • You can regenerate content by starting a new tailoring session.
  • Downloaded DOCX files can be further edited in any word processor.

AI Monitoring

We use Langfuse to monitor AI generation quality, including token usage, latency, and error rates. This monitoring helps us ensure the service works reliably. Monitoring data includes trace IDs and token counts but does not include the content of your CVs or personal information.

Contact

If you have questions or concerns about our use of AI, please contact us at support@wadecv.com.