# 19 · R&D Recipe Sheets · V1 BALANCED + V5 KOJI-MAX + V2 TEMPEH-MAX

> ⛔ **HARD CONSTRAINT — 100% PLANT-BASED**. No egg, no dairy, no animal-derived. All ingredients must be verified halal-compatible (apple cider vinegar, alcohol-free miso, JAKIM-friendly shio-koji) before kitchen-fire.
>
> **For R&D dept**: 3 ready-to-fire kitchen sheets. Hand directly to Tricia / Yivonne / kitchen team. Each sheet has: exact gramage · temperatures · timings · visual cues · verification checkpoints · adjustment ranges (where chef preference flex is OK) · sensory scoring card · failure-mode triage · photo capture list · results template.
>
> Scaled to **400g final batch ≈ 4 portions** (single batch can be re-portioned for tasting panel).

## §0 · Pre-flight checklist (ONE TIME for any of V1/V5/V2)

### Equipment (must be ready before Day-0)

- [ ] Digital probe thermometer (±2°C accuracy) — **mandatory** for steam temp + sear temp
- [ ] Kitchen scale (0.1g precision) — **mandatory** for psyllium %, TGase %, salt
- [ ] pH strips (range 3.0-9.0)
- [ ] Stopwatch / phone timer
- [ ] Plastic pipettes (3-5ml, food-safe) ×5
- [ ] Vacuum sealer (handheld OK) + freezer-zip bags ×20
- [ ] Steamer tray with cover (bamboo or stainless)
- [ ] Cast-iron pan OR heavy-bottom skillet (sear-finish)
- [ ] Mesh strainer + cheesecloth
- [ ] 4× small ramekins for blind tasting
- [ ] Camera (phone OK) + neutral gray plate
- [ ] Printed scoring card (template at §4 of each sheet)

### Sourcing (lead times)

Per `~/.zennith/workspace/intel/texture-engineering-deepdive-2026-05-16/05-SOURCING-MANIFEST.md` consolidated cart. Key items:

| Item | Source | Lead | Cost |
|---|---|---|---|
| Dried hericium | Ganofarm Cameron Highlands | 2-3 day | RM 25-40/kg fresh / RM 80-150/100g dried |
| Vital wheat gluten 500g | Mercato / Lazada | 0-3 day | RM 25-35 |
| Mung bean isolate / Soy isolate / Pea isolate | Bioway | 1-3 day | RM 50-100 each |
| Shio-koji liquid | Don Don Donki KL (Hanamaruki) | 0 day | RM 30-50/300g |
| Tempeh protein powder | Lazada (Hello Tempayy) | 7-14 day | RM 60-90/250g |
| White miso paste (alcohol-free) | Don Don Donki | 0 day | RM 25-40/400g |
| Konjac glucomannan + Ca(OH)₂ | Lazada | 3-14 day | RM 30-50 each |
| Methylcellulose E461 | Lazada / Modernist Pantry | 3-7 day | RM 60-90/250g |
| κ-Carrageenan + alginate + CaCl₂ | Lazada / Modernist Pantry | 3-14 day | RM 60-90/100g each |
| Microbial transglutaminase (Activa-TI) | Lazada (Ajinomoto) | 3-7 day | RM 80-120/100g |
| Apple cider vinegar (Bragg or local halal) | Mercato | 0 day | RM 25/500ml |
| Sea salt (unrefined, Malaysian coastal) | Specialty | 0-7 day | RM 8-15/kg |
| Baking soda (food-grade) | Any supermarket | 0 day | RM 5 |
| Aquafaba (chickpea brine) | FREE — collect from canned chickpeas | 0 day | $0 |
| Amazake | Don Don Donki | 0 day | RM 30/300g |
| Nutritional yeast | Mercato | 0 day | RM 30/250g |
| Kombu (dried strips) | Don Don Donki | 0 day | RM 25/100g |
| Dried shiitake (for powder) | Asian grocer | 0 day | RM 25/100g |

**Sub-total bulk order**: ~RM 600-880 covers 10-20 R&D experiments.

### Halal substitution audit (MANDATORY before Day-0)

| Standard ingredient | Halal-friendly substitute | Why |
|---|---|---|
| Wine vinegar / red wine vinegar | **Apple cider vinegar** (Bragg) | Wine-derived = non-halal |
| Mirin / sake-containing miso | Alcohol-free miso (Marukome Halal-cert) OR house-made | Trace alcohol concern |
| Standard shio-koji | Verify Hanamaruki halal status OR DIY (koji + sea salt + water 3-day ferment, halal-source ingredients) | Some imports use non-halal carriers |

---

# 📋 SHEET 1 · V1 BALANCED (Flagship · Predicted 8.94/10)

**Lane**: FLAGSHIP — mass-market wellness + premium-curious
**Sim endorser**: Hafiz Razak ("The balance of ingredients in V1 is truly remarkable")
**Retail tier**: RM 26-30 per 100g portion
**Predicted composite**: 8.94 / 10
**9-Node scores**: N1=7.7 · N2=9.0 · N3=8.0 · N4=8.0 · X1=10.0 · X2=9.0 · X3=9.0 · X4=9.5 · X5=10.0

## §1.1 · Ingredient table (400g batch = ~4 portions)

| # | Ingredient | Quantity | Tolerance | Function (node) | Supplier |
|---|---|---|---|---|---|
| **SUBSTRATE** | | | | | |
| 1 | Dried hericium | 64g (rehydrates to ~320g) | ±5% | F1 sponge + PMR7 GMP source | Ganofarm |
| **PRE-SOAK** (Step 2) | | | | | |
| 2 | Filtered water (alkali brine) | 2.4L total (in 4 cycles of 600ml) | ±10% | F4 + X1 chitin soften | tap-filtered |
| 3 | Baking soda (NaHCO₃) | 4g | ±0.5g | Alkali → pH 8.5 (X1, X2) | Any grocery |
| 4 | Sea salt (unrefined) | 16g | ±2g | F5 osmosis + bitter suppress | Malaysian coastal |
| **VINEGAR BATH** (Step 5) | | | | | |
| 5 | Apple cider vinegar 5% (Bragg) | 80ml in 1L water | ±10% | F4 chitin hydrolysis (X1) | Mercato |
| **PROTEIN MARINADE** (Step 8 — vacuum-tumble) | | | | | |
| 6 | Aquafaba (chickpea brine) | 100ml | ±20ml | Egg-white mimic (PMR1) | DIY from canned chickpeas |
| 7 | Mung bean protein isolate | 20g | ±2g | Egg-coagulation mimic | Bioway |
| 8 | Vital wheat gluten powder | 40g | ±4g | **F18 disulfide alpha** ⭐ | Mercato |
| 9 | Soy protein isolate (88%+) | 40g | ±4g | F2 + F18 multi-protein gel | Bioway |
| 10 | Methylcellulose E461 | 6g | ±0.5g | **F12 hot-set juicy interior** ⭐ | Lazada |
| 11 | κ-Carrageenan (food-grade) | 2g | ±0.2g | F10 surface skin gel | Lazada |
| 12 | Konjac glucomannan flour | 12g | ±1g | F17 thermo-irreversible firm-bite | Lazada |
| 13 | Microbial transglutaminase (Activa-TI) | 0.25% of total marinade mass ≈ 1g | ±0.1g | F11 covalent cross-link peak | Lazada |
| 14 | Shio-koji liquid (halal-verified) | 40g | ±5g | F3 free Glu liberation | Don Don Donki / DIY |
| 15 | Tempeh protein powder | 40g | ±5g | F3 + 67% protein bioavail | Lazada (Hello Tempayy) |
| 16 | Amazake (alcohol-free) | 32g | ±5g | Reducing sugar for Maillard (N4) | Don Don Donki |
| 17 | Kombu powder (ground) | 4g | ±0.5g | **PMR4 plant Glu alpha** | DDD ground from dried |
| 18 | Dried shiitake powder | 12g | ±1g | **5'-GMP synergy partner** (N2) | DIY ground from dried |
| 19 | Nutritional yeast | 32g | ±3g | Free Glu + Maillard precursor | Mercato |
| 20 | Psyllium husk (food-grade) | 10g | ±0.5g (DO NOT EXCEED 12g — kill zone) | F2 gel-network base | Mercato |
| **SURFACE WASH** (Step 10) | | | | | |
| 21 | Aquafaba (whipped to soft peak) | 60ml | ±10ml | Egg-wash mimic | DIY |
| 22 | Methylcellulose E461 (additional) | 2g | ±0.2g | F12 surface gel | Lazada |
| 23 | κ-Carrageenan (additional) | 1g | ±0.1g | F10 instant skin | Lazada |
| 24 | Nutritional yeast (surface dust) | 6g | ±0.5g | Maillard surface | Mercato |
| **PAN FINISH** (Step 12) | | | | | |
| 25 | Olive oil (extra-virgin) | 30ml | ±5ml | F7 fat carrier (light) | Mercato |
| 26 | Finishing sea salt | 2g | flexible | Final taste | Malaysian coastal |

**Total cost per batch**: ~RM 56-72 (for 4 portions) = **RM 14-18 per 100g portion COGS**

## §1.2 · Step-by-step process (Total time: ~3h 30min)

```
══════════════════════════════════════════════════════════
  V1 BALANCED · MASTER COOK CARD
══════════════════════════════════════════════════════════

T-0 · STEP 1 · WEIGH-OUT (15 min)
  ├─ Lay out all 26 ingredients per §1.1 table
  ├─ Weigh each into labeled dishes/bowls
  ├─ Mark Pre-soak / Vinegar / Marinade / Wash / Finish groups
  └─ Verify pH strips + thermometer working

T+15min · STEP 2 · ALKALI PRE-SOAK (90 min total — 4 cycles)
  ├─ Cycle 1 (15 min):
  │   ├─ Combine 600ml water + 1g baking soda + 4g sea salt
  │   ├─ pH check: must read ~8.5 (use strips)
  │   ├─ Add 16g dried hericium (1/4 of total)
  │   └─ Soak room-temp, cover
  ├─ Repeat 3 more times for remaining hericium
  ├─ At T+105min: total 4 batches × 22min each = 88min effective alkali contact
  └─ END STATE: hericium feels SOFT + GELATINOUS to touch; tripled in size
  ✓ VERIFY: pH strip reads 8.0-8.5 throughout (top off with NaHCO₃ if drops)

T+105min · STEP 3 · PRESS CYCLE 1 (10 min)
  ├─ Drain ALL hericium (keep brine — has alkali residue, compost)
  ├─ Squeeze each piece FIRMLY by hand over a strainer
  ├─ Catch yellow exudate (tyrosinase byproducts — bitter)
  ├─ Rinse pieces in 300ml fresh water briefly
  └─ Press again. Set aside.
  ✓ VERIFY: 3rd rinse water runs CLEAR (means bitter mostly extracted)

T+115min · STEP 4 · PRESS CYCLE 2 (5 min)
  └─ Repeat press + rinse once more (lighter than cycle 1)
  ✓ VERIFY: hericium feels FIRM-DRY (not waterlogged) but still flexible

T+120min · STEP 5 · VINEGAR PRE-TAME (60 min, ROOM TEMP — NO HEAT)
  ├─ Combine 1L water + 80ml apple cider vinegar in bowl
  ├─ pH check: must read ~3.0-3.5
  ├─ Submerge hericium pieces
  └─ Cover. Hold ROOM TEMP × 60 min
  ✓ VERIFY: hericium develops slight tangy smell; texture noticeably softer

T+180min · STEP 6 · PRESS CYCLE 3 (5 min)
  ├─ Drain vinegar bath
  ├─ Press firmly
  ├─ Rinse BRIEFLY in 200ml cool water (don't over-rinse → lose flavor)
  └─ Set aside

T+185min · STEP 7 · SALT-CURE (30 min)
  ├─ Sprinkle 8g sea salt evenly over hericium
  ├─ Let sit room-temp 30 min
  ├─ Water will weep — drain
  └─ Press lightly
  ✓ VERIFY: hericium FIRM + drier; salt-purge water visible

T+215min · STEP 8 · PROTEIN MARINADE PREP (10 min)
  ├─ In large bowl whisk together:
  │   ├─ 100ml aquafaba (slightly foamy)
  │   ├─ 20g mung bean isolate
  │   ├─ 40g vital wheat gluten
  │   ├─ 40g soy isolate
  │   ├─ 6g methylcellulose
  │   ├─ 2g κ-carrageenan
  │   ├─ 12g konjac glucomannan
  │   ├─ 40g shio-koji liquid
  │   ├─ 40g tempeh powder
  │   ├─ 32g amazake
  │   ├─ 4g kombu powder
  │   ├─ 12g shiitake powder
  │   ├─ 32g nutritional yeast
  │   ├─ 10g psyllium husk
  │   └─ Water 200-300ml to make smooth paste
  ├─ Whisk smooth (no lumps; psyllium will thicken)
  ├─ Rest 5 min for hydration
  └─ Add 1g TGase (microbial transglutaminase) → stir gently → use within 2h
  ✓ VERIFY: paste forms soft peak when spoon lifted; smells savory + yeasty

T+225min · STEP 9 · VACUUM-TUMBLE MARINADE (30 min)
  ├─ Combine pressed hericium + ALL marinade in resealable 1L bag
  ├─ Squeeze air out, vacuum-seal
  ├─ Vacuum cycle method (DIY):
  │   ├─ Vacuum-suck → hold 30sec → release → shake 30sec
  │   ├─ Repeat for 30 min (= ~12-15 cycles)
  ├─ Alternative: Instant Pot pressure-cook 5 min @ 100 kPa (high pressure)
  └─ END STATE: hericium pieces HEAVY (~80% weight gain from marinade)
  ✓ VERIFY: marinade penetrates fibers — cross-section shows even saturation

T+255min · STEP 10 · SURFACE EGG-WASH PAINT (5 min)
  ├─ In separate bowl, whip 60ml aquafaba to soft peak (2 min hand-whisk)
  ├─ Fold in 2g methylcellulose + 1g κ-carrageenan + 6g nutritional yeast + pinch salt
  ├─ Mixture: glossy, viscous (thickens fast)
  └─ PAINT each hericium piece on ALL sides with pastry brush
  ✓ VERIFY: ~1mm uniform gel-wash coat on every piece

T+260min · STEP 11 · STEAM-SET (15 min @ 85°C ± 2°C)
  ├─ Bring steamer to 85°C — use probe thermometer touching tray surface
  ├─ Place painted pieces on parchment-lined tray
  ├─ Cover loosely (methylcellulose needs some air)
  ├─ Hold 15 min, CHECK TEMP every 3 min
  ├─ All gels set simultaneously in this window:
  │   ✓ β-conglycinin 7S (soy) — fully denatured 60-75°C
  │   ✓ Mung bean globulin — coagulates 63-93°C
  │   ✓ Aquafaba albumin — coagulates 60-80°C
  │   ✓ Methylcellulose F12 — fibril self-assembly 52-64°C
  │   ✓ κ-Carrageenan + K⁺ — gel sets
  │   ✓ Wheat gluten — partial S-S bond formation
  └─ END STATE: surface FIRM + golden + uniformly set
  ✓ VERIFY: gentle finger-poke → springs back; no wet patches

T+275min · STEP 12 · SEAR-FINISH (3 min @ 180°C)
  ├─ Heat cast-iron to 180°C surface (drop water → instant evaporation)
  ├─ Add 30ml olive oil, swirl
  ├─ Sear 60 sec/side (4 sides for cube; 2 sides for slab)
  ├─ Maillard browning + gluten S-S strengthening
  └─ END STATE: dark caramelized crust, audible sizzle stops, no excess oil pool
  ✓ VERIFY: visible Maillard crust (darker than initial color); juices not weeping

T+278min · STEP 13 · CUT + REST (3 min)
  ├─ Transfer to cutting board
  ├─ Cut WITH spine direction (3-4 cm strips) for lobster-tear anisotropy
  ├─ Rest 3 min (gels fully set on slight cooling)
  └─ Sprinkle finishing sea salt (2g pinch)

T+281min · DONE · Plate

══════════════════════════════════════════════════════════
TOTAL TIME: ~3h 30min
  Hands-on: ~75 min
  Hands-off (soaks): ~2h 15min
══════════════════════════════════════════════════════════
```

## §1.3 · Verification checkpoints (must pass each)

| Step | Visual cue | Probe-test | Pass/fail |
|---|---|---|---|
| 2 (pre-soak) | Hericium tripled in size, gelatinous | pH strip = 8.0-8.5 | If pH <7.5 → add 0.5g NaHCO₃ |
| 3 (press 1) | Yellow exudate visible | 3rd rinse water CLEAR | If water still yellow → 4th cycle |
| 5 (vinegar) | Tangy smell, softer texture | pH = 3.0-3.5 | If pH >4 → add vinegar |
| 7 (salt-cure) | Visible salt-purge water | Hericium firm to touch | If still mushy → 15 more min |
| 8 (marinade) | Soft-peak paste, no lumps | Smells savory | If lumpy → blend smooth |
| 9 (vacuum) | Hericium HEAVY (80% weight gain) | Cross-section shows saturation | If light → 10 more cycles |
| 10 (wash) | 1mm uniform gel-wash | Glossy surface | If dripping → too thin, add 0.5g MC |
| 11 (steam) | Surface firm + golden | Probe 85°C ± 2°C steady | If <80°C → not setting; bump heat |
| 12 (sear) | Visible Maillard crust | Surface 180°C | If pale → 30s more |

## §1.4 · Adjustment ranges (where chef preference flex is OK)

| Parameter | Default | Range | Effect |
|---|---|---|---|
| Press cycles | 3 | 2-4 | More = less umami, less = more bitter risk |
| Vinegar bath time | 60 min | 45-75 min | Longer = softer texture, less acidic flavor |
| Sear temp | 180°C | 170-200°C | Higher = darker crust, risk burn |
| Sear time | 60 sec/side | 45-90 sec | Longer = more Maillard, drier |
| Cut direction | with-spine | with-spine / perpendicular / cube | Affects bite-style only |
| Olive oil amount | 30ml | 20-40ml | Personal fat preference |
| Finishing salt | 2g | 1-3g | Personal salt preference |
| Steam time | 15 min | 12-18 min | Longer = firmer (rubbery risk >20min) |

**DO NOT FLEX** these (chemistry-critical):
- Psyllium dose (10g max — >12g kills gel)
- Konjac (12g — >14g rubbery)
- TGase % (0.25% — peak texturization)
- Methylcellulose dose (6g + 2g surface — F12 sweet spot)
- Pre-soak pH (8.5 — too low = under-softens chitin)
- Steam temp (85°C ± 2°C — protein denaturation window)

## §1.5 · Sensory scoring card (per panelist)

```
════════════════════════════════════════════════════════
  V1 BALANCED · SENSORY PANEL CARD
════════════════════════════════════════════════════════
Date: ________  Panelist: ________  Sample: V1

Score 0-10 each:

  Attribute                | Score | Notes
  ─────────────────────────┼───────┼──────────────────
  1. Umami (depth/persist) |  ___  | ___________
  2. Salty                 |  ___  | ___________
  3. Sweet                 |  ___  | ___________
  4. Sour                  |  ___  | ___________
  5. Bitter (lower=better) |  ___  | ___________
  6. Spicy                 |  ___  | ___________
  7. Aroma intensity       |  ___  | ___________
  8. Texture-bounce        |  ___  | ___________
  9. Bite-through ease     |  ___  | ___________
 10. Aftertaste            |  ___  | ___________
 11. Mealiness (lower=better)|  ___  | ___________
 12. Visual appeal        |  ___  | ___________

PASS GATES:
  ☐ Composite ≥ 8.0 / 10
  ☐ Bite-through ≥ 8 (NOT cloth-like)
  ☐ Bitter ≤ 3
  ☐ Mealiness ≤ 3
  ☐ Halal-compatible verified

Verbatim: _______________________________________
Would buy at RM 26-30? ☐Yes ☐Maybe ☐No
═══════════════════════════════════════════════════════
```

## §1.6 · Failure modes + fixes

| Failure | Likely cause | Fix |
|---|---|---|
| Cloth-texture remaining | Alkali pre-soak too short OR vinegar bath too short | Extend pre-soak to 120min · check baking soda freshness |
| Surface gel detaching during sear | κ-Carrageenan dose too low | Add 0.5g κ-carrageenan to surface wash |
| Mealiness / powdery | Too much isolate, not enough hydration | Add 30ml water in marinade |
| Beany off-note | Pea isolate creep (V1 has none — check sourcing) | Verify ingredient list — V1 has NO pea isolate |
| Rubbery / dense | Over-konjac OR over-steam | Cut konjac to 10g, steam max 15 min |
| Burns before 60s sear | Pan too hot (>220°C) | Drop heat, check thermometer |
| Pieces dry out on cool | Methylcellulose dose too low | Increase total MC to 8g |
| Bitter notes | Press cycles insufficient | 4th press cycle next batch |

## §1.7 · Photo capture (mandatory, all stages)

- [ ] T+0 weigh-out (all ingredients laid out)
- [ ] T+105 alkali pre-soak end (gelatinous hericium)
- [ ] T+115 press cycle 1 (yellow exudate visible)
- [ ] T+180 vinegar bath end
- [ ] T+225 marinade prep (paste consistency)
- [ ] T+255 vacuum-tumble end (cross-section showing saturation)
- [ ] T+275 steam-set end (golden firm surface)
- [ ] T+278 sear-finish (Maillard crust)
- [ ] T+281 cut cross-section (showing internal structure)
- [ ] Final plated dish (top-down + 45° + side)

## §1.8 · Results template (R&D staff fills out)

```yaml
# /Users/jennwoeiloh/.zennith/workspace/intel/texture-engineering-deepdive-2026-05-16/results/V1-BALANCED-result.md
variant: V1_BALANCED
date: 2026-05-XX
chef: <name>
panelists: [<name1>, <name2>, <name3>]
batch_size_g: 400

ingredient_actual_vs_planned: # any deviations
  # e.g. tempeh_powder_g: { planned: 40, actual: 35, reason: "ran low" }

process_deviations:
  # e.g. "steamed at 87°C instead of 85 — couldn't hit exact"

sensory_scores_avg:
  umami: ___
  texture_bounce: ___
  bite_through: ___
  bitter: ___
  composite: ___

photos: [path1, path2, ...]

verdict: PASS | BORDERLINE | FAIL
go_to_menu_launch: YES | NO | REVIEW

next_actions: |
  # what to try next
  # e.g. "increase tempeh to 45g, retest"
```

---

# 📋 SHEET 2 · V5 KOJI-MAX (Ferment-Premium · Predicted 8.72/10)

**Lane**: FERMENT-PREMIUM — fine-dining + fermentation enthusiast
**Sim engagers**: Tricia, Lukas Henning, Chef David Sim
**Retail tier**: RM 32-38 per 100g portion
**Predicted composite**: 8.72 / 10
**9-Node scores**: N1=9.1 · N2=7.9 · N3=8.0 · N4=6.0 · X1=10.0 · X2=9.0 · X3=9.0 · X4=8.0 · X5=10.0

## §2.1 · Ingredient delta vs V1

Same V1 process EXCEPT (per 400g batch):

| Ingredient | V1 | V5 | Δ | Why |
|---|---|---|---|---|
| Shio-koji liquid | 40g | **80g** | +40 | DOUBLE — peak free Glu liberation |
| Soy protein isolate | 40g | **80g** | +40 | MAX protein density |
| Tempeh protein powder | 40g | **100g** | +60 | MAX ferment-protein stack |
| Miso paste (alcohol-free) | 0g | 0g | same | V5 keeps no-miso (lets shio-koji+tempeh dominate) |
| Amazake | 32g | **0g** | -32 | OMIT — reducing-sugar lower for cleaner umami |
| Koji preferment hours | 12h | **12h (refrigerated)** | same | Critical — uncomment if making in advance |
| Sear time | 60 sec | **90 sec/side** | +30 sec | Slightly more Maillard despite no amazake |

**Total cost per V5 batch**: ~RM 78-95 = **RM 19-24 per 100g COGS** (higher tier)

## §2.2 · V5-specific steps (replace Steps 8 + 12 in V1 base)

```
STEP 8b · V5 ENHANCED PROTEIN MARINADE (60 min if pre-fermenting)

OPTION A (recommended): 12h koji preferment
  ├─ At T-12h (day before): mix marinade WITHOUT TGase + hericium pre-soak completed
  ├─ Cover with damp cloth (NOT airtight — koji needs O2)
  ├─ Hold refrigerated 12h (or room temp 8h)
  ├─ Check at 6h: should smell faintly nutty/sweet
  │   ✓ If smells acetone/solvent → STOP, dump
  ├─ At 12h: white mycelium fully covers paste surface, pleasant fruity-nutty smell
  ├─ STOP ferment: cook 5 min at 70°C to kill koji activity
  └─ Proceed with TGase addition + Step 9 vacuum-tumble

OPTION B (faster): no preferment
  └─ Mix all marinade ingredients including 80g shio-koji
  └─ Skip preferment hold, go straight to vacuum-tumble (Step 9)
  └─ Less free Glu liberation but cleaner production timeline

STEP 12b · V5 SEAR-FINISH (90 sec per side)
  ├─ Sear 180°C × 90 sec per side (vs V1's 60 sec)
  └─ Compensates for V5's lack of amazake reducing sugar — slightly longer Maillard
```

## §2.3 · V5 sensory expectations + adjustments

| Aspect | V5 Expected |
|---|---|
| Umami | HIGHER than V1 (9.1 vs 7.7) — peak free Glu |
| Bitter | Slight risk if koji over-fermented (>18h) — kill threshold 4.0/10 |
| Texture | Same as V1 |
| Aroma | Slight "fermenty" note — POSITIVE for ferment enthusiasts, may flag for traditional palates |

**Tricia/Lukas target persona**: high acceptance for funkiness. Auntie Wong target persona: V5 may be TOO funky — give them V1 instead.

## §2.4 · Use V1's scoring card + verification + failure-mode tables

V5 inherits §1.3-1.8 from V1 sheet. Only the ingredient ratios + cook step 8b/12b differ.

---

# 📋 SHEET 3 · V2 TEMPEH-MAX (Protein-Fitness · Predicted 8.85/10)

**Lane**: PROTEIN-FITNESS — James/Hafiz fitness segment
**Retail tier**: RM 28-34 per 100g portion
**Predicted composite**: 8.85 / 10
**9-Node scores**: N1=10.0 · N2=8.4 · N3=8.0 · N4=7.0 · X1=8.5 · X2=9.0 · X3=9.0 · X4=8.5 · X5=10.0

## §3.1 · Ingredient delta vs V1

Same V1 process EXCEPT (per 400g batch):

| Ingredient | V1 | V2 | Δ | Why |
|---|---|---|---|---|
| Alkali pre-soak time | 90 min | **120 min** | +30 | Lower vinegar exposure → more alkali |
| Vinegar bath time | 60 min | **15 min** | -45 | V2 prefers alkali-driven softening |
| Salt-cure time | 30 min | **15 min** | -15 | Lighter cure for protein-density positioning |
| Soy protein isolate | 40g | **60g** | +20 | +50% protein density |
| Pea protein isolate | 0g | **60g** | +60 | NEW — complete amino + iron + WHC |
| Vital wheat gluten | 40g | **40g** | same | F18 alpha preserved |
| Tempeh protein powder | 40g | **100g** | +60 | MAX bioavail protein stack |
| Miso paste | 0g | **40g** | +40 | NEW — long-fermented umami |
| Methylcellulose | 6g | **6g** | same | F12 preserved |
| Aquafaba wash | 60ml | **120ml** | +60 | NEW — surface skin layer (better moisture) |
| Konjac | 12g | **12g** | same | F17 preserved |
| Alginate-Ca | 0g | **2g** | +2 | NEW — F10 surface skin double-stack |
| Sear time | 60 sec | **0 sec** | -60 | V2 SKIPS sear — keeps protein-bar tier mouthfeel |

**Protein content estimate per 100g final**: ~25g protein (vs V1's ~16-18g) — hits HP25 positioning

**Total cost per V2 batch**: ~RM 65-82 = **RM 16-21 per 100g COGS**

## §3.2 · V2-specific steps (replace Step 2 + Step 12 in V1 base)

```
STEP 2b · V2 LONGER ALKALI PRE-SOAK (120 min)
  └─ Same as V1 Step 2 but 4 cycles of 30 min each (total 120 min)
  └─ pH stays at 8.5

STEP 5b · V2 SHORTER VINEGAR BATH (15 min only)
  └─ Submerge hericium in 1L water + 80ml apple cider vinegar
  └─ ROOM TEMP × 15 min (vs V1's 60 min)
  └─ Cellular softening primarily from extended alkali, lighter acid finish

STEP 8b · V2 PROTEIN-MAX MARINADE (10 min prep)
  ├─ Same V1 method but:
  │   ├─ Soy isolate 60g (vs 40g)
  │   ├─ Pea isolate 60g (NEW)
  │   ├─ Tempeh powder 100g (vs 40g)
  │   ├─ Miso paste 40g (NEW)
  │   └─ Optional: roast pea isolate 3 min in dry pan @ 120°C BEFORE adding
  │     (reduces beany off-note via early Maillard)
  └─ Add TGase 0.25% (≈ 1.5g for larger marinade mass) → rest 5 min

STEP 10b · V2 DOUBLE SURFACE WASH (5 min)
  ├─ Standard V1 Step 10 wash
  ├─ THEN pre-dip pieces in calcium chloride bath:
  │   ├─ 1L water + 5g CaCl₂
  │   ├─ Dip 5 sec each piece (alginate-Ca surface skin forms instantly)
  └─ Then standard V1 paint procedure

STEP 12b · V2 NO-SEAR FINISH (steam-set only)
  ├─ After Step 11 steam-set, REMOVE FROM HEAT immediately
  ├─ Glaze with finishing salt + small drizzle olive oil
  ├─ Skip cast-iron sear entirely
  └─ Result: clean protein-bar-tier mouthfeel without Maillard crust
  ✓ POSITIONING: "protein-density bar" not "meat-mimic" — different sensory class
```

## §3.3 · V2 sensory expectations + adjustments

| Aspect | V2 Expected |
|---|---|
| **Protein density** | **HIGHEST** (~25g/100g) — fitness-segment alpha |
| Umami | Top score (10/10) — kombu + miso + tempeh + nut yeast stack |
| Beany off-note | Risk from pea isolate — MITIGATE via pre-roast (Step 8b) |
| Maillard depth | LOW (no sear) — POSITIONING choice for clean-protein bar segment |
| Texture | Firmer than V1 due to higher isolate density |
| Cost-per-gram-protein | LOWEST of 3 variants — value win for HP customer |

## §3.4 · Use V1's scoring card + verification + failure-mode tables

V2 inherits §1.3-1.8 from V1 sheet. Only the ingredient ratios + cook step 2b/5b/8b/10b/12b differ.

---

# 📋 R&D Department Master Sheet

## §M.1 · Suggested fire order

| Day | Activity | Cost | Effort |
|---|---|---|---|
| Day 0 | Equipment + sourcing pre-flight check | 0 | 1h |
| Day 1 | BASELINE: cook V1 alone (no comparison) | RM 56 | 4h |
| Day 2 | Day-2 Panel: V1 alone, 4-5 tasters, lock baseline | RM 50 | 2h |
| Day 3 | V5 KOJI-MAX with 12h preferment day-before | RM 90 | 4h |
| Day 4 | V5 panel + V1 comparison | RM 0 | 2h |
| Day 5 | V2 TEMPEH-MAX | RM 75 | 3.5h |
| Day 6 | V2 panel + 3-way comparison V1/V5/V2 | RM 50 | 2h |
| Day 7 | Synthesis + menu launch decision | 0 | 2h |

**Total program**: ~RM 320 budget · ~20h labor across 7 days

## §M.2 · Cross-variant scoring matrix (for Day 6 comparison)

| Attribute | V1 BAL | V5 KOJI | V2 TEMPEH | Best variant |
|---|---|---|---|---|
| Umami | _.__ | _.__ | _.__ | __ |
| Texture bounce | _.__ | _.__ | _.__ | __ |
| Bite-through ease | _.__ | _.__ | _.__ | __ |
| Aroma intensity | _.__ | _.__ | _.__ | __ |
| Aftertaste | _.__ | _.__ | _.__ | __ |
| Cost per 100g COGS | RM 14-18 | RM 19-24 | RM 16-21 | __ |
| Persona fit | broad | premium-fine-dining | fitness-HP | __ |
| Composite | _.__ | _.__ | _.__ | __ |

## §M.3 · Questions for R&D dept to verify

Hand-list for kitchen team:

1. **Halal cert of Hanamaruki shio-koji** — verify before Day 1 OR substitute house-made
2. **Apple cider vinegar source** — Bragg verified halal? Or local halal-cert?
3. **TGase Activa-TI** — confirm Ajinomoto halal status (typically yes)
4. **Tempeh powder Hello Tempayy** — confirm 24h-fermented (not 36h+ over-ferment)
5. **Hericium grade** — fresh ≥80g pieces preferred for spine-cut anisotropy
6. **Aquafaba source** — DIY from boiled chickpeas (cleaner) OR canned (faster — verify BPA-free)
7. **Steam tray temp accuracy** — pre-test thermometer placement
8. **Cast-iron pre-heat** — at least 10 min before Step 12 sear
9. **Press cycle force** — moderate squeeze (not maximum — preserves anisotropy)
10. **Storage between batches** — vacuum-pack + refrigerate ≤24h if making in advance

## §M.4 · Scaling guide (for production)

| Batch | Multiply all ingredients by | Yields |
|---|---|---|
| Test (R&D) | 1× | 4 portions |
| Pinxin pilot | 25× | 100 portions |
| Pinxin Week-1 production | 250× | 1000 portions |
| Full ramp | 2500× | 10,000 portions/week |

Pinxin existing -40°C blast-freeze infrastructure handles scaling. NO new CapEx.