diff --git a/documentation/pipeline/design-doc.md b/documentation/pipeline/design-doc.md
index a1d7666..255c04e 100644
--- a/documentation/pipeline/design-doc.md
+++ b/documentation/pipeline/design-doc.md
@@ -3,7 +3,7 @@
 > **Project:** PERN-stack vocabulary trainer with Gemini-powered data pipeline
 > **Author:** [Your Name]
 > **Date:** July 2026
-> **Status:** Approved — ready for implementation
+> **Status:** Approved — schema implemented
 
 ---
 
@@ -83,9 +83,8 @@ staging).
 │ headword    │  └───>│ word_id(FK) │  └───>│ sense_id (FK)    │
 │ language_code│      │ sense_index │       │ target_lang_code │
 │ pos         │       │ difficulty  │       │ translation      │
-│             │       │ cefr_level  │       │ gender           │
-│             │       │ definitions │       │ difficulty       │
-│             │       │ examples    │       │                  │
+│             │       │ definitions │       │ gender           │
+│             │       │ examples    │       │ difficulty       │
 └─────────────┘       └─────────────┘       └──────────────────┘
 
 Future (not yet implemented):
@@ -119,7 +118,7 @@ One row per unique word in a specific language.
 
 **Index:**
 
-- `idx_words_lang_pos ON (language_code, pos)` — accelerates the
+- `idx_language_code_pos ON (language_code, pos)` — accelerates the
   primary game query filter.
 
 **Design note:** Each language gets its own headword entries. "Haus"
@@ -139,7 +138,6 @@ two senses of one word.
 | word_id     | UUID        | FK → words.id, ON DELETE CASCADE |                                              |
 | sense_index | SMALLINT    | NOT NULL, default 0              | 0 = primary meaning, 1 = secondary, etc.     |
 | difficulty  | VARCHAR(20) | NOT NULL, CHECK in allowed list  | "easy", "medium", "hard"                     |
-| cefr_level  | VARCHAR(2)  | nullable, CHECK in allowed list  | "A1","A2","B1","B2","C1","C2"                |
 | definitions | TEXT[]      | NOT NULL, default '{}'           | 1–3 definitions in the word's language       |
 | examples    | TEXT[]      | NOT NULL, default '{}'           | 1–3 example sentences in the word's language |
 | created_at  | TIMESTAMPTZ | NOT NULL, default now()          |                                              |
@@ -148,18 +146,11 @@ two senses of one word.
 
 - `UNIQUE (word_id, sense_index)` — one sense per index per word.
 - `CHECK (difficulty IN ('easy','medium','hard'))`
-- `CHECK (cefr_level IS NULL OR cefr_level IN ('A1','A2','B1','B2','C1','C2'))`
-
-**CEFR → difficulty mapping:**
-
-- A1, A2 → easy
-- B1, B2 → medium
-- C1, C2 → hard
 
 **Index:**
 
-- `idx_senses_word_diff ON (word_id, difficulty)` — accelerates the
-  join from words and the difficulty filter.
+- `idx_word_sense_difficulty ON (word_id, difficulty)` — accelerates
+  the join from words and the difficulty filter.
 
 **Design note — definitions and examples as arrays:**
 Definitions and examples are stored as `TEXT[]` arrays on the sense
@@ -201,7 +192,7 @@ different difficulty levels (e.g., "Bank" easy, "Geldinstitut" medium).
 
 **Index:**
 
-- `idx_translations_sense_lang_diff ON (sense_id, target_language_code, difficulty)`
+- `idx_translations_sense_language_difficulty ON (sense_id, target_language_code, difficulty)`
   — accelerates the join from senses and the language/difficulty filter.
 
 **Design note — gender as a real column:**
@@ -409,7 +400,6 @@ The API is prompted to return an array of objects. Expected shape:
     "senses": [
       {
         "sense_index": 0,
-        "cefr_level": "A1",
         "difficulty": "easy",
         "definitions": ["Ein Gebäude zum Wohnen."],
         "examples": ["Sie kauften ein Haus in der Stadt."],
@@ -442,7 +432,6 @@ The API is prompted to return an array of objects. Expected shape:
       },
       {
         "sense_index": 1,
-        "cefr_level": "C1",
         "difficulty": "hard",
         "definitions": ["Ein Adelsgeschlecht, eine Dynastie."],
         "examples": ["Das Haus der Merowinger herrschte über Franken."],
@@ -474,8 +463,6 @@ Before writing to SQLite, every entry is checked:
 - At least one sense per word.
 - Each sense has at least one definition and one example.
 - `difficulty` is one of: `easy`, `medium`, `hard`.
-- `cefr_level` is one of: `A1`, `A2`, `B1`, `B2`, `C1`, `C2`.
-- CEFR → difficulty mapping is consistent.
 - `gender` is valid for the target language:
   - German: masculine, feminine, neuter
   - French, Spanish, Italian: masculine, feminine
@@ -502,13 +489,13 @@ used only as a pipeline staging file.
 Three indexes cover the game and distractor queries:
 
 ```sql
-CREATE INDEX idx_words_lang_pos
+CREATE INDEX idx_language_code_pos
   ON words (language_code, pos);
 
-CREATE INDEX idx_senses_word_diff
+CREATE INDEX idx_word_sense_difficulty
   ON senses (word_id, difficulty);
 
-CREATE INDEX idx_translations_sense_lang_diff
+CREATE INDEX idx_translations_sense_language_difficulty
   ON translations (sense_id, target_language_code, difficulty);
 ```
 
@@ -563,13 +550,13 @@ one-time or occasional operation.
 ## 9. Implementation Plan
 
 ```
- 1. ✅ Design doc (this document)
- 2.    Acquire word frequency lists (5 languages, nouns first)
- 3.    Build + test Gemini prompt (5 words → 20 words → full batch)
- 4.    Validation script (Gemini output → clean JSON)
- 5.    SQLite staging schema + pipeline write
- 6.    Import script (SQLite → local Postgres)
- 7.    Drizzle schema: words, senses, translations + indexes
+ 1. ✅ Schema design (this document)
+ 2. ✅ Drizzle schema: words, senses, translations, relations, indexes
+ 3.    Acquire word frequency lists (5 languages, nouns first)
+ 4.    Build + test Gemini prompt (5 words → 20 words → full batch)
+ 5.    Validation script (Gemini output → clean JSON)
+ 6.    SQLite staging schema + pipeline write
+ 7.    Import script (SQLite → local Postgres)
  8.    Drizzle migration on local Postgres
  9.    Dev branch: new queries (game + distractor), full game flow test
 10.    Drizzle migration on prod Postgres + data import + verify
diff --git a/documentation/pipeline/roadmap.md b/documentation/pipeline/roadmap.md
index c6baef7..f01bcb5 100644
--- a/documentation/pipeline/roadmap.md
+++ b/documentation/pipeline/roadmap.md
@@ -14,42 +14,48 @@
 
 This roadmap has three zoom levels:
 
-- **Part 1 — Overview:** The phases at a glance. Read this to understand the full scope in 30 seconds.
-- **Part 2 — Phase Breakdown:** Goals, tasks, dependencies, and acceptance criteria per phase. Read this to plan your week.
-- **Part 3 — Detailed Tasks:** Step-by-step instructions within each phase. Read this when you sit down to code.
+- **Part 1 — Overview:** The phases at a glance. Read this to
+  understand the full scope in 30 seconds.
+- **Part 2 — Phase Breakdown:** Goals, tasks, dependencies, and
+  acceptance criteria per phase. Read this to plan your week.
+- **Part 3 — Detailed Tasks:** Step-by-step instructions within each
+  phase. Read this when you sit down to code.
 
 ---
 
 # Part 1 — High-Level Overview
 
 ```
-Phase 0  Preparation
+Phase 1  Schema ✅
+         New Drizzle schema (words, senses, translations) with
+         relations, constraints, and indexes. Committed.
+
+Phase 2  Preparation
          Get wordlists, set up tooling, finalize the Gemini prompt.
 
-Phase 1  Data Pipeline
+Phase 3  Data Pipeline
          Build the Gemini → validate → SQLite pipeline.
          Produce a clean dataset of ~1000 nouns × 5 languages.
 
-Phase 2  Schema & Migration
-         Create the new Drizzle schema (words, senses, translations).
+Phase 4  Migration & Import
+         Generate and apply the Drizzle migration.
          Write the SQLite → Postgres import script.
-         Migrate the local dev database.
 
-Phase 3  App Integration
+Phase 5  App Integration
          Rewrite the game and distractor queries against the new schema.
          Update the exercise-generation logic.
          Test the full game flow in dev.
 
-Phase 4  Production Deploy
+Phase 6  Production Deploy
          Run the Drizzle migration on prod.
          Import the dataset.
          Verify the live app works end-to-end.
 
-Phase 5  Extend POS
+Phase 7  Extend POS
          Run the pipeline for verbs, adjectives, adverbs.
          No schema changes needed — new wordlists + adjusted prompts.
 
-Phase 6  Future Features (out of scope for now)
+Phase 8  Future Features (out of scope for now)
          Inflection tables, conjugation/declension exercises,
          gender exercises, spaced-repetition scheduling.
 ```
@@ -57,22 +63,41 @@ Phase 6  Future Features (out of scope for now)
 **Dependency chain:**
 
 ```
-Phase 0 → Phase 1 → Phase 2 → Phase 3 → Phase 4 → Phase 5
-                                                      │
-                                                      ▼
-                                                  Phase 6
+Phase 1 ✅ → Phase 2 → Phase 3 → Phase 4 → Phase 5 → Phase 6 → Phase 7
+                                                                    │
+                                                                    ▼
+                                                                Phase 8
 ```
 
-Each phase depends on the previous one being complete. Do not skip
-ahead.
-
 ---
 
 # Part 2 — Phase Breakdown
 
 ---
 
-## Phase 0: Preparation
+## Phase 1: Schema ✅ COMPLETE
+
+**Goal:** New normalized schema exists in Drizzle with all tables,
+constraints, indexes, and relations.
+
+**Completed tasks:**
+
+- [x] `words` table: headword, language_code, pos, UNIQUE, CHECKs, index
+- [x] `senses` table: word_id FK (cascade), sense_index, difficulty,
+      definitions TEXT[], examples TEXT[], UNIQUE, CHECK, index
+- [x] `translations` table: sense_id FK (cascade), target_language_code,
+      translation, gender (nullable), difficulty, UNIQUE, CHECKs, index
+- [x] Relations: words→senses (many), senses→word (one) + translations
+      (many), translations→sense (one)
+- [x] `NOUN_GENDERS` constant added to `@lila/shared`
+- [x] `DIFFICULTY_LEVELS` updated: "intermediate" → "medium"
+- [x] Old tables (`vocabulary_entries`, `entry_translations`) untouched
+- [x] Auth and lobby tables untouched
+- [x] Build passes, committed
+
+---
+
+## Phase 2: Preparation
 
 **Goal:** Have everything you need before writing pipeline code.
 
@@ -86,11 +111,12 @@ ahead.
 - [ ] Refine the prompt until the JSON output matches the contract
       defined in `docs/schema-design.md` §6.3
 
-**Dependencies:** None.
+**Dependencies:** Phase 1 complete.
 
 **Acceptance criteria:**
 
-- You have 5 wordlist files in `data/wordlists/` (one per language).
+- You have 5 wordlist files in `data-pipeline/source-data/` (one per
+  language).
 - A Gemini call with 5 German nouns returns valid JSON matching the
   contract, including definitions, examples, translations with gender,
   and difficulty levels.
@@ -98,7 +124,7 @@ ahead.
 
 ---
 
-## Phase 1: Data Pipeline
+## Phase 3: Data Pipeline
 
 **Goal:** A repeatable script that takes a wordlist, calls Gemini in
 batches of 20, validates the output, and writes clean rows to SQLite.
@@ -112,7 +138,7 @@ batches of 20, validates the output, and writes clean rows to SQLite.
 - [ ] Review the rejection log, fix prompt issues, re-run failed batches
 - [ ] Spot-check 50 random entries for correctness
 
-**Dependencies:** Phase 0 complete.
+**Dependencies:** Phase 2 complete.
 
 **Acceptance criteria:**
 
@@ -125,16 +151,13 @@ batches of 20, validates the output, and writes clean rows to SQLite.
 
 ---
 
-## Phase 2: Schema & Migration
+## Phase 4: Migration & Import
 
-**Goal:** The new normalized schema exists in Drizzle, the local
-Postgres database is migrated, and the SQLite data is imported.
+**Goal:** The new schema exists in Postgres and the SQLite data is
+imported.
 
 **Tasks:**
 
-- [ ] Write the new Drizzle schema file: - `words` table (headword, language_code, pos) - `senses` table (word_id FK, sense_index, difficulty,
-      cefr_level, definitions TEXT[], examples TEXT[]) - `translations` table (sense_id FK, target_language_code,
-      translation, gender, difficulty) - All CHECK constraints, UNIQUE constraints, indexes
 - [ ] Generate the Drizzle migration (`npx drizzle-kit generate`)
 - [ ] Inspect the generated SQL file — verify it creates the right
       tables, constraints, and indexes
@@ -146,7 +169,7 @@ Postgres database is migrated, and the SQLite data is imported.
 - [ ] Run 3–5 manual SQL queries against Postgres to sanity-check
       the data
 
-**Dependencies:** Phase 1 complete (SQLite has data).
+**Dependencies:** Phase 3 complete (SQLite has data).
 
 **Acceptance criteria:**
 
@@ -158,7 +181,7 @@ Postgres database is migrated, and the SQLite data is imported.
 
 ---
 
-## Phase 3: App Integration
+## Phase 5: App Integration
 
 **Goal:** The running app uses the new schema. Game rounds and
 distractors work correctly for all language pairs.
@@ -177,7 +200,7 @@ distractors work correctly for all language pairs.
 - [ ] Test edge cases: - A difficulty/pos/language combo with very few words
       (does the app handle < 4 available words gracefully?) - A word with multiple senses (does the correct sense appear?)
 
-**Dependencies:** Phase 2 complete (Postgres has data, schema exists).
+**Dependencies:** Phase 4 complete (Postgres has data, schema exists).
 
 **Acceptance criteria:**
 
@@ -189,7 +212,7 @@ distractors work correctly for all language pairs.
 
 ---
 
-## Phase 4: Production Deploy
+## Phase 6: Production Deploy
 
 **Goal:** The live app runs on the new schema with the new data.
 
@@ -203,7 +226,7 @@ distractors work correctly for all language pairs.
 - [ ] Monitor for errors (server logs, browser console) for 24h
 - [ ] Remove old tables from prod (after confirming everything works)
 
-**Dependencies:** Phase 3 complete (dev is fully working).
+**Dependencies:** Phase 5 complete (dev is fully working).
 
 **Acceptance criteria:**
 
@@ -214,7 +237,7 @@ distractors work correctly for all language pairs.
 
 ---
 
-## Phase 5: Extend POS
+## Phase 7: Extend POS
 
 **Goal:** Verbs, adjectives, and adverbs are in the database and
 usable in the app.
@@ -229,7 +252,7 @@ usable in the app.
 - [ ] Test game flow with verbs, adjectives, adverbs
 - [ ] Verify the POS filter in the app UI works for all types
 
-**Dependencies:** Phase 4 complete.
+**Dependencies:** Phase 6 complete.
 
 **Acceptance criteria:**
 
@@ -238,7 +261,7 @@ usable in the app.
 
 ---
 
-## Phase 6: Future Features (out of scope)
+## Phase 8: Future Features (out of scope)
 
 Listed here for visibility. Not planned, not estimated.
 
@@ -255,9 +278,9 @@ Listed here for visibility. Not planned, not estimated.
 
 ---
 
-## Phase 0: Preparation — Detailed
+## Phase 2: Preparation — Detailed
 
-### 0.1 Acquire wordlists
+### 2.1 Acquire wordlists
 
 - Search for frequency lists. Good starting points:
   - "Leipzig Corpora Collection" (academic, per-language)
@@ -267,11 +290,11 @@ Listed here for visibility. Not planned, not estimated.
   - Tatoeba sentence counts as a proxy for word frequency
 - Target: ~1000 nouns per language, sorted by frequency.
 - Format: plain text, one word per line, UTF-8, no headers.
-- Save to `data/wordlists/nouns_de.txt`, `nouns_es.txt`, etc.
+- Save to `data-pipeline/source-data/{language}/nouns/`.
 - Clean the lists: remove duplicates, remove words with spaces
   (multi-word expressions), remove proper nouns if desired.
 
-### 0.2 Set up local Postgres
+### 2.2 Set up local Postgres
 
 - Option A (Docker):
   ```
@@ -288,7 +311,7 @@ Listed here for visibility. Not planned, not estimated.
 - Verify: connect with `psql` or a GUI client (TablePlus, DBeaver,
   pgAdmin). Run `SELECT 1;`.
 
-### 0.3 Write and test the Gemini prompt
+### 2.3 Write and test the Gemini prompt
 
 - Start with 5 German nouns.
 - The prompt should specify:
@@ -297,7 +320,6 @@ Listed here for visibility. Not planned, not estimated.
   - That translations are needed for all 4 other supported languages
   - That gender must be provided for de/fr/es/it, null for en
   - That difficulty must be one of: easy, medium, hard
-  - That CEFR level must be one of: A1, A2, B1, B2, C1, C2
   - That multiple senses should be included for polysemous words
 - Call the API. Inspect the JSON. Common issues to fix:
   - Gemini wraps the JSON in markdown code fences → strip them
@@ -308,12 +330,12 @@ Listed here for visibility. Not planned, not estimated.
 
 ---
 
-## Phase 1: Data Pipeline — Detailed
+## Phase 3: Data Pipeline — Detailed
 
-### 1.1 Create the SQLite schema
+### 3.1 Create the SQLite schema
 
-- Create a file `pipeline/schema.sql` or define it in your pipeline
-  script.
+- Create a file `data-pipeline/schema.sql` or define it in your
+  pipeline script.
 - Tables mirror the Postgres schema:
 
   ```sql
@@ -331,7 +353,6 @@ Listed here for visibility. Not planned, not estimated.
     word_id TEXT NOT NULL REFERENCES words(id),
     sense_index INTEGER NOT NULL DEFAULT 0,
     difficulty TEXT NOT NULL,
-    cefr_level TEXT,
     definitions TEXT NOT NULL DEFAULT '[]',  -- JSON array as text
     examples TEXT NOT NULL DEFAULT '[]',     -- JSON array as text
     created_at TEXT DEFAULT (datetime('now')),
@@ -353,9 +374,9 @@ Listed here for visibility. Not planned, not estimated.
 - Note: SQLite has no native TEXT[]. Store arrays as JSON text.
   The import script will parse them into Postgres TEXT[] on import.
 
-### 1.2 Write the validation module
+### 3.2 Write the validation module
 
-- Create `pipeline/validate.ts` (or .js).
+- Create `data-pipeline/validate.ts`.
 - Input: one parsed Gemini entry (the JSON object for one word).
 - Checks (return a list of errors, empty = valid):
   - `headword` is a non-empty string
@@ -365,8 +386,6 @@ Listed here for visibility. Not planned, not estimated.
   - Each sense has:
     - `sense_index` is a non-negative integer
     - `difficulty` is in ['easy','medium','hard']
-    - `cefr_level` is in ['A1','A2','B1','B2','C1','C2'] or null
-    - CEFR ↔ difficulty mapping is consistent
     - `definitions` is a non-empty array of non-empty strings
     - `examples` is a non-empty array of non-empty strings
     - `translations` is a non-empty array
@@ -381,9 +400,9 @@ Listed here for visibility. Not planned, not estimated.
       - `difficulty` is in ['easy','medium','hard']
 - Output: `{ valid: boolean, errors: string[] }`
 
-### 1.3 Write the pipeline script
+### 3.3 Write the pipeline script
 
-- Create `pipeline/run.ts`.
+- Create `data-pipeline/run.ts`.
 - Pseudocode:
   ```
   for each language in [de, en, es, fr, it]:
@@ -398,7 +417,7 @@ Listed here for visibility. Not planned, not estimated.
           generate UUIDs for word, senses, translations
           INSERT into SQLite (words, senses, translations)
         else:
-          append to rejection log (data/rejections/{lang}_{pos}.jsonl)
+          append to rejection log
       log progress: "Batch 12/50 done. 238 words imported, 2 rejected."
       sleep 1s (rate limiting)
   ```
@@ -407,7 +426,7 @@ Listed here for visibility. Not planned, not estimated.
 - Store definitions/examples as JSON strings in SQLite
   (`JSON.stringify(arr)`).
 
-### 1.4 Run and review
+### 3.4 Run and review
 
 - Run the pipeline for all 5 languages.
 - Check the rejection log. If rejection rate > 10%, fix the prompt
@@ -427,31 +446,21 @@ Listed here for visibility. Not planned, not estimated.
 
 ---
 
-## Phase 2: Schema & Migration — Detailed
+## Phase 4: Migration & Import — Detailed
 
-### 2.1 Write the Drizzle schema
-
-- Create or update `src/db/schema.ts`.
-- Define the three tables using Drizzle's `pgTable` builder.
-- Include all columns, types, constraints, and indexes as specified
-  in `docs/schema-design.md` §3.
-- Keep the old tables in the file for now (commented out or in a
-  separate file) so the app doesn't break during development.
-
-### 2.2 Generate and inspect the migration
+### 4.1 Generate and inspect the migration
 
 - Run: `npx drizzle-kit generate`
-- Open the generated SQL file in `drizzle/` (or wherever your
-  config puts it).
+- Open the generated SQL file in `packages/db/drizzle/`.
 - Read it. Verify:
-  - Three CREATE TABLE statements
+  - Three CREATE TABLE statements (words, senses, translations)
   - CHECK constraints match your schema
   - UNIQUE constraints are present
   - Three CREATE INDEX statements
   - Foreign keys reference the correct tables with ON DELETE CASCADE
 - If something looks wrong, fix the schema file and regenerate.
 
-### 2.3 Apply the migration locally
+### 4.2 Apply the migration locally
 
 - Run: `npx drizzle-kit migrate`
 - Connect to local Postgres and verify:
@@ -462,9 +471,9 @@ Listed here for visibility. Not planned, not estimated.
   \d translations              -- describe translations table
   ```
 
-### 2.4 Write the import script
+### 4.3 Write the import script
 
-- Create `scripts/import-to-postgres.ts`.
+- Create `data-pipeline/import-to-postgres.ts`.
 - Pseudocode:
 
   ```
@@ -492,7 +501,7 @@ Listed here for visibility. Not planned, not estimated.
   into actual arrays (Postgres TEXT[]).
 - Use `ON CONFLICT DO NOTHING` to handle re-runs gracefully.
 
-### 2.5 Run and verify
+### 4.4 Run and verify
 
 - Run the import script.
 - Compare counts:
@@ -529,11 +538,11 @@ Listed here for visibility. Not planned, not estimated.
 
 ---
 
-## Phase 3: App Integration — Detailed
+## Phase 5: App Integration — Detailed
 
-### 3.1 Rewrite getGameTerms
+### 5.1 Rewrite getGameTerms
 
-- Open the file containing `getGameTerms`.
+- Open `packages/db/src/models/termModel.ts`.
 - Replace the old query (2-table join on `vocabulary_entries` +
   `entry_translations`) with the new 3-table join
   (words → senses → translations).
@@ -545,7 +554,7 @@ Listed here for visibility. Not planned, not estimated.
 - Return: word_id, headword, sense_id, definitions, examples,
   translation, gender.
 
-### 3.2 Rewrite getDistractors
+### 5.2 Rewrite getDistractors
 
 - Same 3-table join.
 - Additional filters:
@@ -555,7 +564,7 @@ Listed here for visibility. Not planned, not estimated.
 - If fewer than 3 distractors are found (small data pool), handle
   gracefully: reduce the number of options or log a warning.
 
-### 3.3 Update exercise generation
+### 5.3 Update exercise generation
 
 - In the function that assembles a game round:
   - Pick one random definition:
@@ -566,7 +575,7 @@ Listed here for visibility. Not planned, not estimated.
   - Shuffle the 4 options (Fisher-Yates or similar).
   - Attach gender to each option for display.
 
-### 3.4 Test matrix
+### 5.4 Test matrix
 
 Run through this matrix manually in the dev app:
 
@@ -584,18 +593,18 @@ appear as distractors, no duplicate options.
 
 ---
 
-## Phase 4: Production Deploy — Detailed
+## Phase 6: Production Deploy — Detailed
 
-### 4.1 Pre-deploy checklist
+### 6.1 Pre-deploy checklist
 
-- [ ] All Phase 3 acceptance criteria pass
+- [ ] All Phase 5 acceptance criteria pass
 - [ ] `git status` is clean, all changes committed
 - [ ] The Drizzle migration file is committed to the repo
 - [ ] You know your prod database connection string
 - [ ] You have a backup method for prod (pg_dump, hosting provider
       snapshot, etc.)
 
-### 4.2 Deploy
+### 6.2 Deploy
 
 - Back up prod:
   ```
@@ -607,11 +616,11 @@ appear as distractors, no duplicate options.
   ```
 - Run import script against prod:
   ```
-  DATABASE_URL=<prod-url> npx tsx scripts/import-to-postgres.ts
+  DATABASE_URL=<prod-url> npx tsx data-pipeline/import-to-postgres.ts
   ```
 - Verify counts on prod.
 
-### 4.3 Post-deploy verification
+### 6.3 Post-deploy verification
 
 - Open the live app. Play 2 full games with different settings.
 - Check server logs for errors.
@@ -623,24 +632,24 @@ appear as distractors, no duplicate options.
   ```
   (Or keep them for another week if you want a safety net.)
 
-### 4.4 Rollback plan
+### 6.4 Rollback plan
 
 If something goes wrong:
 
 - Restore the backup:
   ```
-  psql -h <host> -U <user> -d <db> < backup_20260722.sql
+  psql -h <host> -U <user> -d <db> < backup_YYYYMMDD.sql
   ```
 - Revert the code to the previous commit.
 - Redeploy.
 
 ---
 
-## Phase 5: Extend POS — Detailed
+## Phase 7: Extend POS — Detailed
 
-### 5.1 Per POS
+### 7.1 Per POS
 
-Repeat the Phase 1 pipeline for each new POS:
+Repeat the Phase 3 pipeline for each new POS:
 
 - Acquire wordlists (verbs, adjectives, adverbs) for all 5 languages.
 - Adjust the Gemini prompt:
@@ -653,7 +662,7 @@ Repeat the Phase 1 pipeline for each new POS:
 - Run pipeline → validate → SQLite → import → Postgres.
 - Test in the app with the POS filter set to the new type.
 
-### 5.2 No schema changes
+### 7.2 No schema changes
 
 The `pos` column already supports all four types. The `gender`
 column is nullable and simply won't be populated for adverbs or
@@ -678,27 +687,27 @@ English words. No migration needed.
 
 ```
 project/
-├── docs/
-│   ├── schema-design.md          ← companion design doc
-│   └── roadmap.md                ← this document
-├── data/
-│   ├── wordlists/
-│   │   ├── nouns_de.txt
-│   │   ├── nouns_en.txt
-│   │   ├── nouns_es.txt
-│   │   ├── nouns_fr.txt
-│   │   └── nouns_it.txt
+├── data-pipeline/
+│   ├── source-data/
+│   │   ├── german/nouns/         ← wordlist files
+│   │   ├── english/nouns/
+│   │   ├── spanish/nouns/
+│   │   ├── french/nouns/
+│   │   └── italian/nouns/
 │   ├── rejections/               ← invalid Gemini entries (for review)
-│   └── staging.db                ← SQLite staging database
-├── pipeline/
-│   ├── run.ts                    ← main pipeline script
+│   ├── staging.db                ← SQLite staging database
+│   ├── schema.sql                ← SQLite schema definition
 │   ├── validate.ts               ← validation module
-│   ├── prompt.ts                 ← Gemini prompt template
-│   └── schema.sql                ← SQLite schema definition
-├── scripts/
+│   ├── run.ts                    ← main pipeline script
 │   └── import-to-postgres.ts     ← SQLite → Postgres import
-├── src/
-│   └── db/
-│       └── schema.ts             ← Drizzle schema (updated)
-└── drizzle/                      ← generated migration files
+├── packages/
+│   ├── db/
+│   │   └── src/db/schema.ts      ← Drizzle schema (updated ✅)
+│   └── shared/
+│       └── src/constants.ts      ← NOUN_GENDERS added ✅, "medium" ✅
+├── documentation/
+│   └── pipeline/
+│       ├── design-doc.md         ← schema design doc (updated ✅)
+│       └── roadmap.md            ← this document (updated ✅)
+└── ...
 ```