Export Application

Export Application

Produce the final export after all answers are approved and scored >= 6.0.

---

Output 1 — Review Copy (a16z-application-draft.md)

Structure:

# a16z Speedrun Application — [Company Name]
Generated: [date]
Eval score: [average score across all fields]

---

## [Field name]
**Character limit:** [limit if known]
**Score:** [X.X/10]

[Answer text]

---

[repeat for each field]

Save to: ./a16z-application-[company-slug]-[YYYY-MM-DD].md

---

Output 2 — Copy-Paste Sheet (a16z-copypaste.txt)

Plain text, no markdown. Each field block:

=== [FIELD NAME] ===
[Answer text — plain text only, no special chars that could break form inputs]

[blank line between fields]

Save to: ./a16z-copypaste-[company-slug]-[YYYY-MM-DD].txt

---

Output 3 — Version Record (a16z-application.json)

For tracking across application iterations:

{
  "company": "[name]",
  "generated": "[ISO date]",
  "version": 1,
  "eval_average": [score],
  "fields": [
    {
      "field": "[field name]",
      "answer": "[answer text]",
      "score": [score],
      "char_count": [count]
    }
  ]
}

Save to: ./a16z-application-[company-slug].json (overwrite on re-export, bump version integer).

---

Post-export checklist

Display after export:

EXPORT COMPLETE

Files written:
  a16z-application-[slug]-[date].md     (review copy)
  a16z-copypaste-[slug]-[date].txt      (paste into form)
  a16z-application-[slug].json          (version record)

BEFORE YOU SUBMIT:
  [ ] Read every answer out loud. Fix anything that sounds like marketing copy.
  [ ] Verify every number is accurate. Do not submit estimates as facts.
  [ ] Check character limits on each field against the live form.
  [ ] Have one other person read the "why you" and "what's your insight" fields.
  [ ] Record your video pitch AFTER reading the script 3+ times. Cut anything that sounds rehearsed.

a16z reviews on a rolling basis. Submit now — do not wait for the "perfect" version.