Job-search English Evidence Card: From Job Description to Unfamiliar Follow-up

This card does not decide whether you are suited to overseas work or promise an interview result. It answers a narrower question: under this role and process, can I deliver my real ability in English?

Copy it into a private note. Job descriptions, recruiting email, take-home tasks, company code, customer information, and interview recordings may contain confidential or personal information. Preserve only the minimum material you have a right to use, and obtain necessary consent before recording.

1. Role and Process

Role title and public link:
Application deadline or interview date:
Work mode: remote / hybrid / on-site
Primary time zones and collaboration mode:
Known interview stages:
Explicit language requirement:
Recruiting rules still unknown:

2. Role Language Map

Role fact	Language task	Current evidence	Risk	Build evidence in two weeks?
	recruiter communication / project explanation / technical follow-up / behavioural question / async writing / customer communication		high / medium / low

Can complete independently:
Can complete with preparation:
Depends on prompts or translation:
Has no sample:
One highest-risk gap to repair:

3. Complete Baseline

Date:
Preparation time:
Visible support: none / outline / full script
Tools: none / dictionary / captions / AI / other
Recording and text location:

Stage	Time	Acceptance	Independent result	Main error
Role-relevant introduction	2 min	Present work, evidence, direction
Project explanation	5 min	Context, responsibility, trade-off, result, unknown
Unfamiliar follow-up	3 questions	Understand, answer, or clarify
Asynchronous update	20 min	Progress, evidence, risk, next step, owner
Interaction repair	1 case	Repeat/confirm/thinking time/unknown boundary

4. Error Map

Segment/question	Type	Effect on the task	What to test next
	content / retrieval / listening / organisation / interaction / accuracy

Largest problem that is not primarily English:
Language error that repeatedly changes key meaning:
Repair action worth preserving:

5. Real Story Bank

Keep six lines rather than a full script:

Story name:
Situation and required task:
My specific responsibility:
Key action and trade-off:
Result, cost, and unknown:
What I would change now:
Evidence location:

Story	Questions it can answer	Available after unfamiliar follow-up?	Fact/privacy check
	problem / failure / conflict / decision / feedback / stopping	yes / no

6. Listening and Repair

Original question I did not understand:
Sentence used to request repetition or confirmation:
Meaning after restatement:
My answer:
How shared understanding was confirmed:

Repair ability	Baseline 0–2	Day seven 0–2	Evidence
Request a specific repetition
Confirm scope
Ask for thinking time
Admit an unknown and give a verification step
Summarise agreement and next action

7. Asynchronous Writing Sample

Context:
Completed:
Evidence:
Blocked / risk:
Decision needed:
Owner and next update:

Ask one reader:

What happened:
What needs a decision:
Who continues and when:
Sentence that required guessing:

8. AI and Integrity Record

Tool and date:
Role it simulated:
Independent version submitted first:
Three problems it raised:
Feedback accepted/rejected and why:
Parallel result after closing the tool:
Stage where AI must be disclosed or is prohibited:
Confidential/personal information kept out:

Do not invent experience, numbers, or ownership. Do not use covert real-time prompting in a live interview without explicit permission.

9. Fourteen-Day Comparison

Time point	Condition	Task completion	Understanding/repair	Organisation/retrieval	Async writing	Evidence
Day-one baseline	Unscripted, no AI
Immediate revision	Feedback seen
Day-seven retest	Parallel questions, original answer removed
Day-fourteen transfer	New interviewer/order/role question

Score each 0–2: 0=not completed, 1=partial or prompt-dependent, 2=completed under the current condition. Use only for one person's comparable tasks.

10. Current Role Gates

• [ ] At least three real stories remain available under changed wording.
• [ ] Recruiter screen, project explanation, and unfamiliar follow-up can be completed.
• [ ] Unclear questions produce clarification rather than guessing.
• [ ] Repeated language errors do not change key meaning.
• [ ] One usable asynchronous update can be written.
• [ ] Confidentiality, recording, and AI boundaries are known.
• [ ] Unknowns can be admitted with a verification method.
• [ ] Equipment, time zone, and interview environment are checked.

Current evidence most supports: apply and continue practising / repair one high-risk gap / defer this role
Evidence:
Conclusion I still cannot draw:
Smallest next task:
Next review date:

11. Post-interview Close

Actual questions:
Evidence of correct understanding:
Clarification that worked:
Story that failed to answer the question:
Structure that transferred:
Difficulty not caused primarily by English:
Recruiter feedback in original wording:
One variable to change:
Preparation to stop:

Related chapters: Job-search English: Bring Ability into Interviews and Remote Work | Speaking Evidence Card | Writing Evidence Card