skills/lq-apply/references/formats.md

# The three formats

## 1. The LQ application

The real application lives on legalquants.com; its questions are the target.
Question wording is owned by the site's `lib/apply/questions.ts` — read the
live wording from the site when drafting (never hardcode labels from memory;
they were finalized 2026-07-15 and may evolve). When the site is unreachable,
say the draft is provisional: never claim a fixed question count or an exact
form match. The mapping below is stable by slug.

Draftable from the journey and records today:

| Slug | Drafted from |
|---|---|
| `practice-where` | practice.sentence |
| `ai-journey` | fluency history + graduated lessons (the growth narrative: where they started, what changed) |
| `ai-failure` | friction events — the lesson, and how they dug out (graduated status is the digging) |
| `lq-moment` | kept lq_moment events — pick the strongest: what + technique |
| `ai-edge` | the moments' technique tags — the reproducible edge, stated concretely |
| `ai-setup` | technique tags + tools seen across debriefs |
| `ai-usage` | debrief/moment cadence as evidence of sustained use — and note the user supplies the provider screenshot themselves |

Asked inline, marked as gaps (the rest, today): `urls-built`, `most-used`,
`own-initiative`, `most-paid`, `week-one-plan`, `opportunities`,
`how-heard`. These are not in the profile by design — ask each plainly,
one at a time, and place their answers verbatim.

The honest account: a count of what was drafted from evidence
versus what needs them ("X answers drafted from your journey and records; Y
need you") + the one line per gap that maps to future evidence. When the
live form was unreachable, the count refers to the last known mapping and
says so.

## 2. The CV (one page)

- **Header:** name placeholder + practice sentence.
- **The edge:** 2–3 lines from technique tags — what they can do that an
  equally good lawyer without the setup can't.
- **Selected achievements:** 3–5 kept moments as outcome bullets: the kind
  of task, the technique, the counterfactual ("a morning's cross-document
  redline pass, delivered before the first meeting").
- **Growth:** graduated lessons, one line each — proof the practice
  improves.
- **Cadence:** sustained-use evidence ("weekly reviewed AI practice since
  <month>", from debrief counts) — no self-assessed skill ratings.

## 3. The public profile draft

- **Tagline:** one line, practice + edge, no buzzwords.
- **Bio paragraph:** practice sentence woven with the journey's arc.
- **Known-for entries:** 3–5, built from technique tags (label = the
  technique family, value = one concrete line).
- **Flagship:** the single strongest kept moment, told in two sentences
  with its counterfactual.

Voice for all three: evidence, never adjectives; outcomes, never tool
name-dropping; no verdicts, no self-ratings, no "passionate about".