Explore

Farin

1) UI Quality

Expectation

Match the actual intended UI experience from Figma (visual look, spacing rhythm, interaction state), not only copied numeric values.

If design interpretation is unclear, ask before marking complete.

Actions

Before development: review design and write a short “UI interpretation note” (2–4 bullets).

Before handoff: run a visual parity checklist (desktop + at least one other viewport).

Attach comparison evidence (screenshots/GIF).

Success criteria

No “looks off” feedback in first review for assigned work.

At least 90% UI parity accepted in first pass.

2) UX Decision Quality

Expectation

Validate whether flow feels usable and trustworthy for a client.

If behavior is unclear, clarify with lead/QA before implementation.

Actions

For each feature, define:

expected user path,

edge behavior (empty/error/loading),

success/failure messaging.

Confirm behavior with QA/lead when ambiguous.

Success criteria

No major UX behavior mismatch after “done” handoff.

All ambiguous behavior clarified before implementation.

3) Status Update Accuracy

Expectation

Report actual status honestly; never call “done” if only partial.

No stakeholder update that overstates readiness.

Actions

Use fixed status labels:

In Progress

Ready for Internal Review

Ready for QA

Every update must include:

completed scope,

pending scope,

known bugs/risks.

Success criteria

Zero “done but not done” incidents.

Zero incorrect “main flow done” reports.

4) Rework Reduction / Scope Discipline

Expectation

Do not modify restricted areas unless explicitly approved.

Preserve previously fixed/approved sections.

Actions

Before coding, list in-scope files/modules.

Any out-of-scope change requires approval note.

If adjacent issues are found, log separately instead of silently modifying.

Success criteria

Zero unauthorized scope touches.

Rework requests due to unnecessary edits reduced significantly.

5) Testing Discipline

Expectation

Properly test own code before handoff (not just AI-generated output submission).

Verify with different screens/data states.

Actions

Mandatory self-test checklist per task:

happy path,

at least 2 edge cases,

cross-screen/responsive check,

regression check on touched modules.

Include test evidence in handoff (brief checklist + proof).

Success criteria

Majority of defects found before handoff (self-detected).

Critical bugs in first review drop sharply.

6) Debugging & Problem-Finding First

Expectation

Identify root cause before proposing fixes.

Use GPT/Copilot as support, not substitute for diagnosis.

Actions

For each bug, submit mini-debug note:

symptom,

root cause hypothesis,

validation steps,

chosen fix + why.

Escalate only after own investigation attempt.

Success criteria

Clear root-cause reasoning shown on assigned bugs.

Improved independence in troubleshooting.

7) Guidance Follow-through (remote + office)

Expectation

Apply guidance consistently in both offline and remote work.

Practice assigned debugging techniques.

Actions

Weekly check-in on:

what guidance was applied,

what was learned,

what still blocks execution.

Maintain short “learning log” (debug skills practiced this week).

Success criteria

Demonstrable use of suggested debugging methods.

Fewer repeated mistakes across similar tasks.

Want to print your doc?
This is not the way.

Try clicking the ··· in the right corner or using a keyboard shortcut (

CtrlP

) instead.