Job search
Testlify AI Interview: Review the Language, Recording, and Submission Instructions
Identify Phone Call, Voice Web or Chat Web, review language and recording instructions, and distinguish conversation end from final submission confirmation.
Before a Testlify AI interview, identify whether the invitation uses Phone (Call), Voice (Web) or Chat (Web), then check the assigned language, recording notice and final submission steps. The end of a conversation does not by itself confirm assessment submission. In the documented phone flow, “Call Completed” is followed by a separate Submit action.
This guide reviews official public documentation checked October 23, 2026. We opened no private invitation, granted no recording permissions and made no call, recording or assessment submission. Both worked cases are original fiction. Keep the actual employer instructions beside your preparation record; for separate applications, browse recent roles on LandOffer.ai.

Identify the interview mode before choosing your setup
Testlify’s AI interview setup guide, updated July 11, 2026, names three modes. Phone uses an AI call; Voice Web uses spoken interaction in the browser; Chat Web is the text-based route. This is employer-facing configuration documentation; it describes available modes rather than proving which one your invitation assigns.
A phone interview can be created separately or embedded in an assessment. Its presence therefore does not tell you whether coding, qualifying questions or other sections remain. Record each component shown in the genuine invitation rather than assuming the AI conversation is the entire application stage.
The Voice AI guide, updated July 10, 2026, describes browser-based responses requiring a microphone and a call option requiring a telephone number. A browser microphone check and the ability to receive a telephone call solve different setup needs. Neither establishes that your employer has enabled both options or lets you switch between them.
Identify whether the task is a conversational AI interview, an ordinary recorded response or a read-aloud feature. Similar equipment does not make the workflows interchangeable. If your invitation says only “AI interview,” ask which response mode and which other components apply before starting.
Confirm the language and what will be captured
Testlify’s complete candidate guide, updated July 11, 2026, says assessments use the language selected by the employer and directs candidates to the employer for a language accommodation. A translated browser menu or your account’s preferred language does not establish that the interview accepts responses in that language.
Record the required response language explicitly. If it is missing or inconsistent with the role’s instructions, ask before beginning. Do not rely on a general multilingual marketing claim as permission to change languages during a particular assessment. A requested change remains a request until the responsible party confirms the arrangement.
The Voice AI guide says spoken responses are recorded and available in the employer’s candidate results. The phone guide describes call recordings, transcripts and scores in the employer view. Review the notice attached to your own process before supplying a number, speaking or granting browser permissions. A quiet screen does not imply the response is unrecorded.
The provider’s enterprise data reference also lists snapshots, video and audio among data collected according to assessment configuration. That is a reason to inspect the applicable notice, not a claim that every Chat Web interview records a webcam or that every candidate has the same configuration.
Ask what is captured, who can review it, how it is used, how long it is retained and where process-specific questions should go. We did not inspect an employer’s notice or verify an individual retention period. A recording setting, browser permission and informed understanding of the notice are separate items in your preparation record.
Completed mode audit: one invitation, three possible assumptions
Fictional case: Kira receives an invented invitation for a software-support role at Alder Bridge Services. It names Voice (Web) in English. She initially considers preparing for a telephone call because another application used phone screening. No real invitation, phone number or candidate device is used here.
She completes this comparison against the mode actually named in the hypothetical invitation:
| Possible route | Setup implication from the public description | Kira’s completed finding |
|---|---|---|
| Phone (Call) | Candidate supplies a number through the designated flow and receives an AI call | Not the named route; do not wait for an unsolicited call or provide a number elsewhere |
| Voice (Web) | Spoken browser interaction; microphone and a stable connection needed | Matches the fictional invitation; plan the browser audio setup |
| Chat (Web) | Text-based AI conversation in the browser | Not the named route; do not assume typed responses can replace the spoken requirement |
| Video or proctoring component | Camera requirements depend on the actual assessment components | Invitation excerpt does not settle this; ask rather than assume camera-free access |
| Language | Employer-selected language identified as English | Preserve English as assigned; any change needs clarification |
| Recording notice | No process-specific notice included in the fictional excerpt | Locate and review it before beginning; do not mark reviewed |
| Completion | Final screens not supplied | Keep submission instructions unresolved rather than invent a button sequence |
Her completed preparation note reads: “Use Voice Web as the planning route. Do not convert it into Phone or Chat based on earlier applications. Confirm whether other components require a camera, review the applicable recording notice and locate the instructions for ending and submitting the session. No permissions, recording or assessment actions have occurred.”
Kira now knows which preparation question matters. She does not need a working telephone call setup merely because Call mode exists. She does need to know whether a mixed assessment introduces camera requirements beyond Voice Web. Her mode is matched, while the record keeps camera requirements, notice review and completion instructions open for clarification.
Her next-contact note is precise: “Ask the coordinator identified in the original invitation about the missing notice, any additional camera-enabled components and final completion evidence.” It contains no actual email address, verification code or assessment URL. A public help article can inform the question; it cannot supply the missing employer instruction.
Distinguish conversational AI from ordinary recorded responses
Testlify’s September 11, 2026, AI prompt guide describes employer-configured Basic and Advanced prompt approaches. One uses specified questions; the other gives broader conversation instructions. The page notes that configured time can limit how many specified questions are asked. Do not treat a personal rehearsal script as the exact conversation you will encounter.
Prepare truthful explanations of your own work and listen to the actual question. You can organize project context, your contribution, decisions and limitations beforehand without asking another system to impersonate you or produce responses during a restricted assessment. Keep client or employer secrets out of examples unless you are authorized to share them.
Ordinary video questions have their own controls. Testlify’s video-question guide, updated July 13, 2026, says the employer chooses recording attempts and that re-recording is available before submission when multiple attempts are allowed. That does not grant every conversational AI interview a replay, editing pass or fresh attempt after submission.
The separate Video AI creation guide describes an AI interviewer and optional configuration controls. The presence of an avatar, transcript or AI score does not establish identical controls across video-related products. Use the actual component name and displayed instructions to identify what you may do.
Confirm the final submission instead of relying on silence
In the documented Call mode, the candidate reaches a question page, provides the designated contact number, initiates the call and returns to a page showing Call Completed. The guide then specifies Submit to finish. A disconnected line, a silent handset or an elapsed conversation is not equivalent to that final action and its resulting status.
For Voice Web and Chat Web, read the displayed completion instructions for the assigned component. We did not verify their live buttons or a universal final-screen sequence. Record the confirmation text and which component it covers; do not import Call mode’s exact labels into every mode.
An individual component may also sit within a larger assessment. Keep “AI interview step complete,” “all required components complete” and “employer application decision” as separate states. The candidate guide describes a confirmation message at assessment completion and says results go to the employer; neither establishes that an applicant has passed or been selected.

Independent Call-mode case: completion evidence is missing
Separate fictional case: Emil’s invented application at Birch Harbor Tools uses Phone (Call). Its stipulated instructions specify English and include a recording notice, which the case assumes he reviewed. A later scenario describes a call ending, but provides no verified final submission. This is administrative fiction, not a call or test we performed.
| Record item | Supplied hypothetical observation | Completed interpretation |
|---|---|---|
| Mode and language | Phone (Call), English | Keep these attached to this application only |
| Call interval | November 24, 2026, 11:10–11:19 a.m. Eastern Standard Time | Nine-minute invented interval; not a Testlify default duration |
| Page after the call | Scenario says Call Completed appeared | Conversation step reached its documented end state |
| Submit action | No action or resulting state supplied | Final submission remains unconfirmed |
| Other components | Two additional components named in the fictional welcome record; their status absent | Do not mark the whole assessment complete |
| Invitation deadline | November 24 at 1 p.m. EST | Preserve deadline; no extension assumed |
| Retake permission | Not supplied | Do not restart the call or use a second identity |
Emil’s completed incident note is: “Call ended; Call Completed stipulated; final submission and remaining components not established. Consult the current instructions and ask the administrator to confirm recorded status if the page or next step is unclear. No retake permission or completion confirmation has been observed.”
His full unsent clarification reads:
To: Assessment contact from the original invitation, role placeholder only
Subject: Phone interview step — request to confirm submission status
Hello,
My preparation record identifies the Phone Call interview in English. In the scenario I am documenting, the call ended at 11:19 a.m. Eastern Standard Time on November 24 and the page displayed Call Completed. I do not have a confirmed Submit action or final assessment-completion message in the record.
Could you confirm whether the interview response was received and advise which permitted step remains? The welcome record also names two additional components, whose completion I cannot establish. The invitation deadline is 1 p.m. EST that day; I have not assumed an extension or a new attempt.
Before any authorized new interview attempt, please clarify the applicable recording notice and any language or response-mode changes. I can provide the relevant error, browser and device details through the approved support channel without sharing answers, verification codes or a full invitation link.
Thank you,
Emil Example
The record remains submission unconfirmed, with a named administrative next step. The nine-minute call interval and the 101 minutes between 11:19 and 1 p.m. describe only the invented timeline. They do not establish remaining test time, successful upload or how quickly support will respond. No message was sent and no response was invented.
Keep retries, breaks and technical help within their stated scope
The candidate guide says to contact the employer if a retake is needed. Distinguish re-recording before submission where a specific video question permits it from reopening an invitation, repeating a call or obtaining another assessment attempt. None should be inferred from the existence of a public demonstration or employer configuration option.
A pause between ordinary assessment sections is also different from stopping a live AI conversation. Do not assume that closing the browser or ending a call preserves a pause state. For a technical problem, record the mode, time and last confirmed stage, and use the support route provided for that assessment. Questions about language arrangements, attempts and hiring results belong with the responsible employer.
Save a completion record: mode confirmed, language known, notice reviewed, required setup checked and final state observed or unresolved. Then return to your broader search through recent roles on LandOffer.ai, keeping interview recordings, private invitations and candidate answers outside public job-search notes.
Sources and Further Reading
Official public sources checked October 23, 2026. Employer configuration descriptions were reviewed, without candidate-account access, recording permissions, calls, answers or submissions.
- Testlify: Phone, Voice and Chat AI interview modes, July 11, 2026.
- Testlify: Voice AI question modes and recording, July 10, 2026.
- Testlify: complete candidate guide, July 11, 2026.
- Testlify: enterprise data reference, July 13, 2026.
- Testlify: Basic and Advanced AI prompt configuration, September 11, 2026.
- Testlify: ordinary video questions and recording attempts, July 13, 2026.
- Testlify: Video AI custom questions, July 6, 2026.


