1. Information we process
CVeat (cveat.in) provides CV building, ATS analysis, cover letter generation and conversational interview practice. Depending on the features you choose, we process account identifiers and contact details; profile and career information; CVs, job descriptions and uploaded documents; interview questions, typed answers, transcripts and feedback; payment and credit records; and technical information used to operate and protect the service.
Microphone recordings are processed only when you enable the microphone and accept the recording notice. Interview practice does not access or record your camera. Please avoid submitting identity numbers, banking details, health information, confidential employer material or other unnecessary sensitive data.
2. Why we process information
We use this information to maintain your account and drafts, create and export CVs, analyse job matches, generate cover letters and application email drafts, prepare interview questions, transcribe spoken answers, produce practice feedback, manage payments and credits, prevent abuse, troubleshoot failures and respond to support requests. AI feedback is practice guidance, not an employer hiring decision.
The company registration form can be filled before sign-in. Its draft is retained in the current browser tab for up to 24 hours to support sign-in and reloads, and is cleared after successful submission. A verified account is required to submit, and company workspace access requires review. CVeat administrators may email a company registration link using our Hostinger mailbox. We record the recipient address, company/contact name, send attempt and mail-server acceptance status to manage invitations and avoid duplicate messages. Related invitation records are linked to the company when its owner registers with the same verified email and are removed with that company’s live workspace deletion. Invitations sent to a different address may need separate identification for a deletion request. Company workspaces also collect business identity, authorized representative and billing details, including a GSTIN where provided. These are private to authorized company settings users and platform administrators. Employer AI tools send the selected JD and recruiter-provided instructions or interview notes to Gemini to draft job descriptions, questions, factual interview summaries and candidate messages. Generated drafts are saved with the job; raw tool instructions and pasted interview notes are not separately saved by CVeat. Recruiters must have authority to share this information, check generated drafts and send candidate messages themselves. Company deletion also removes these registration details and saved drafts from the live service. When you submit a job application, we share the selected saved-CV text copy or uploaded PDF, your name and verified account email with authorised reviewers at that employer. Uploaded application PDFs are stored privately. An authorised employer reviewer may request Gemini-generated factual CV summaries against that job description. This sends the selected application CV to Google Gemini and stores the resulting evidence summary with the application for authorised company reviewers. Summaries may contain mistakes and do not automatically score, rank, reject or shortlist applicants. Authorised employer reviewers can save manual applicant priorities and internal review notes, which are private to authorised company reviewers. AI does not assign these priorities. The Gemini data-use terms described below also apply to this processing. Submitted CV copies are separate from My History, so editing or deleting a saved CV does not change an application already submitted. Application records and uploaded files are retained for recruitment and support handling; contact support@cveat.in for correction, withdrawal or deletion requests. An authorised platform owner can permanently delete a company workspace and its application files from the live service. Separate user accounts and their personal CVs are outside a company-only deletion. Usage records can be retained with user and request links removed, along with a deletion receipt without company identifiers. Provider backups, provider-held processing logs and copies previously downloaded by recipients may remain subject to separate retention and handling. Using our CV or interview tools does not by itself publish your CV to recruiters. If you choose to share a CV, submit an application or follow an employer’s external link, review the recipient’s privacy terms before sending information. We do not sell personal CV records to data brokers.
3. AI providers, including Sarvam
Sarvam AI — spoken questions: When voice playback is enabled, CVeat sends the interview question text to Sarvam’s text-to-speech API to generate synthetic speech. Questions can contain information derived from your CV or target role. The current integration does not send your recorded answers or full CV file directly to Sarvam and does not clone your voice.
Sarvam may process submitted text, generated audio and associated request or service logs. Retention and any model-improvement use depend on the applicable agreement and account settings; we do not promise zero retention or no training without verifying those settings. Read Sarvam’s Privacy Policy, EULA and API retention documentation.
Google Gemini — AI analysis and transcription: CV content, job descriptions and answers are sent to Gemini as needed for the selected AI feature. Recordings may be sent for transcription when you choose that mode or browser transcription is unavailable. CVeat uses Gemini API through a Google Cloud project with active billing. Under Google’s Paid Services terms, submitted prompts, files and responses are not used to improve Google’s products. Google processes this content under its applicable data processing terms and may retain prompts and responses for a limited period for policy enforcement, security and required legal disclosures. This does not mean zero retention. Account, billing and technical usage data are handled under Google’s applicable privacy terms. See the Gemini API terms. Share only information needed for the selected feature and that you are authorised to provide.
Browser speech recognition: Optional live transcription may send speech to the speech service provided by your browser. That provider’s processing terms apply. You can use typed answers instead of the microphone, turn voice playback off, and revoke microphone permission in your browser.
4. Hosting, authentication and payments
Supabase supports account authentication and database storage. Our hosting and infrastructure services process requests and technical logs. Razorpay processes checkout and payment details; CVeat retains transaction identifiers, payment status, billing details and credit records needed for fulfilment and support, rather than asking you to enter payment-card credentials into the CV editor. Email and support delivery services may process messages you send or receive.
Providers may process information in countries where they or their subprocessors operate. Storage location and safeguards depend on the specific service and account configuration; CVeat does not make a blanket promise that every provider processes all data only in India.
5. Browser storage and cookies
We use browser local storage and session storage for authentication sessions, CV drafts and selected workflow transfers. Browser storage is not automatically encrypted by CVeat and can remain on a shared device. Sign out after use and clear site storage when appropriate; clearing it can remove unsaved drafts.
For live visitor statistics, a first-party random browser identifier is stored locally for up to 24 hours. Foreground pages send a presence update about once per minute; we store a server-hashed identifier, guest/signed-in category and last-active timestamp, without storing an account ID, page history or IP address in this presence table. Counts cover the last five minutes; records older than 15 minutes are cleaned up on the next presence update. Admin pages and browsers signalling Do Not Track are excluded. Clearing site storage resets the browser identifier. Counts are approximate and can omit browsers that block storage or requests. Essential storage supports the features you request. Any introduction of optional advertising or analytics tracking needs a corresponding notice and consent controls where required. Third-party sites you visit through links have their own cookie practices.
6. Retention, recordings and deletion
In personal interview practice, audio clips used for playback and answer review are held temporarily in the interview tab and are not saved to Supabase Storage by the current interview flow. This does not mean the transcription provider retains no data. Closing the tab ends access to these temporary clips unless you downloaded them yourself.
CV drafts remain in browser storage until cleared or replaced. To vary interview practice, recent question text is stored on the same browser for up to 30 days, with a maximum of 30 recent question sets shared across accounts on that browser. Questions are matched separately by account and target role; CVs, job descriptions, answers and audio are not included in this record. Expired entries are ignored and removed when a new set is recorded. Clearing site data removes this record. Questions in your saved Interview History can also be used to avoid repeats for the same role across devices. Selecting Repeat an interview reuses its exact questions with fresh answers. To preserve interview setup through sign-in, the role, job description and selected CV file can be kept in browser IndexedDB. This setup draft is removed after questions are successfully prepared; drafts expire after 30 minutes and expired copies are removed when the draft store is next accessed. You can also clear site data to remove them. Account information, saved CVs, evaluations, operational interview records, AI usage records and payment records are separate categories. Operational records can exist even if you do not choose Save to History. Not all server records currently have an automatic deletion schedule.
History limits are shown in the product; replacing an oldest save requires confirmation. Deleting or replacing a History item does not automatically erase separate billing records, security logs, provider-held copies or backups. To request broader deletion or an explanation of retention, contact support@cveat.in. Some records may need to be retained for legal, security or payment obligations; backup deletion follows the relevant provider’s retention cycle.
For cover letter generation, your uploaded CV, JD and selected details are sent to Google Gemini after your confirmation. The upload is processed in memory by CVeat and is not stored in Supabase Storage by this tool. A draft, including the selected CV file, can be retained in browser IndexedDB for up to 30 minutes to support reload and sign-in; expired local drafts are removed when the draft store is next accessed. Use Clear draft to remove the current local copy. Generated letters and their display details are privately retained for request recovery for 24 hours. Expired server letter text is cleared when that record is next accessed; this is not a scheduled deletion process. Credit and AI usage records are retained separately. The tool does not send emails or submit applications. Review generated content before using it.
Employer-assigned recorded interviews
Employer-assigned interviews are separate from personal interview practice. Before starting, candidates see the employer, recording requirements and retention information and must agree to camera and microphone recording. Video with audio, a separate compressed audio copy for transcription and generated question audio are stored in a private Cloudflare R2 bucket. Supabase stores the application-linked interview record, assigned question and job snapshots, consent timestamp, answer timings, optional written clarifications, transcripts and factual summaries. Only the application account can take its interview, and authorised employer reviewers and platform administrators can review it.
Recording uploads occur during answers. Temporary pieces may remain in browser IndexedDB until uploaded and confirmed; they are scoped to the signed-in account and interview. Keep the tab open until submission is confirmed. Reopening the same invitation in the same browser may recover remaining pieces, but recovery cannot restore cleared or unavailable local storage. Clear site data on shared devices after a confirmed submission. The browser logs limited technical events such as changing tabs, leaving fullscreen, device interruption and connection loss for human review. These events do not prove cheating, and CVeat does not analyse appearance, emotion or personality to make hiring decisions.
Employer reviewers may authorise Gemini transcription of recorded answer audio and factual summaries of transcripts. Sarvam receives question text for synthetic speech; candidate recordings are not sent to Sarvam by this integration. AI outputs require human verification. HR decides recruitment outcomes. Private playback links expire after ten minutes. Recording access ends 30 days after interview start; scheduled cleanup and bucket lifecycle rules remove the cloud recordings, with deletion timing dependent on those services. Transcripts, summaries and recruitment records remain with the application until separately removed. Account, application or company deletion queues associated recording files for removal; queued cleanup can require retries. Contact support@cveat.in about withdrawal, correction or deletion, including employer-held downloads and provider backups.
7. Your choices, rights and security
You can edit profile information, delete saved History items, clear local drafts, use typed answers and stop microphone use. Contact support@cveat.in for access, correction, account deletion, consent withdrawal or a privacy complaint. We may need to verify account ownership. Applicable legal exceptions and required retention can limit deletion; withdrawing consent can make dependent features unavailable.
To reduce repeated free-credit signups, we require a mobile number and check for duplicate numbers and known temporary email domains. Mobile numbers are not currently verified by OTP. New accounts must confirm their email before using AI features. We use a random browser identifier in local storage and server-keyed hashes of network addresses and browser identifiers for signup rate limits. Rate-limit records expire after at most 24 hours and are removed on subsequent signup checks. One-use signup tickets expire after 10 minutes and are removed on subsequent checks. Mobile-account claims are retained separately from editable profile details to prevent repeat trial claims, including after account deletion, unless removed following a support review. Contact support@cveat.in to correct a wrongly linked number or discuss deletion. We use measures such as encrypted HTTPS connections and access controls. No service can guarantee absolute security. AI features are intended for adults aged 18 or over. If information about a child has been submitted, contact us to request review and removal.
8. Changes and contact
Check the date above for policy changes. Material changes should be explained through an appropriate service notice, with fresh consent where required. Privacy enquiries and complaints can be sent to support@cveat.in.