Hypothesis
Retrospective review expectation stated at recording time: the complete recorded
formulation and Method can reach its intended endpoints and produce the following
finished dish: Caramel custard ice cream layered with a soft, salty caramel ripple. This is not a hypothesis documented before historical
cooking and is not itself an observed outcome.
Procedure
This record was written on 2026-10-02 from Carlo Kruger’s retrospective report
on that date. Historical cooking dates and trial counts were not recorded.
The exact whole Recipe Version is Salted Caramel Ice Cream with Salted Caramel Ripple v1.
Its preserved source is recipes-tmp/salted-caramel-ice-cream-with-salted-caramel-ripple.md with SHA-256 8e481cd3b6957e72f322956cae4befa25149c0c96f293b9fe9dc89f79868ed6b.
The serialized draft SHA-256 is cd8da536475b467d44a4acd845c42586e6f310ecdd3c23066e2b854c6a2647b9.
The complete formulation and work sequence below are the recorded specification,
not newly observed timings, temperatures or yields. Correspondence to past cooking
is attributed to Carlo’s statement that the recipes as recorded were fully tested.
The historical reconstruction caveats remain in the unchanged source; Carlo’s
2026-10-02 acceptance applies to this formulation and the packet’s explicit clarifications.
PHASE A — CARAMELISE SUGAR FOR BASE
Ingredient Uses
Method
- Heat 110 g sucrose in a heavy saucepan to a dark amber caramel.
- Carefully add 180 g warm cream in stages. Stir until smooth, then add milk.
Phase Outputs
| Key |
Phase Output |
Description |
| caramel-dairy-base |
Caramel dairy base |
Prepared intermediate for a later Phase. Use all. |
PHASE B — MIX AND PASTEURISE BASE
Ingredient Uses
Method
- Whisk dextrose, skim milk powder, stabiliser, and salt together. Blend into the caramel dairy mixture.
- Temper egg yolk with the warm mixture, recombine, and heat while stirring to 82 C–84 C for 30 s.
- Blend for 30 s and cool to below 4 C as quickly as possible.
- Refrigerate for 6 h–12 h.
Phase Outputs Used
| Phase Output |
Use |
| Caramel dairy base |
Use caramel-dairy-base. Use all. |
Phase Outputs
| Key |
Phase Output |
Description |
| aged-base |
Aged base |
Prepared intermediate for a later Phase. Churn all. |
PHASE C — CARAMEL RIPPLE
Ingredient Uses
| Key |
Ingredient |
Quantity |
Scaling |
Use |
| sucrose |
Sucrose |
150 g |
50.00% |
Sucrose. |
| glucose-syrup |
Glucose syrup |
50 g |
16.67% |
Glucose syrup. |
| cream-warm |
Cream, warm |
150 g |
50.00% |
Cream, warm. |
| unsalted-butter |
Unsalted butter |
40 g |
13.33% |
Unsalted butter. |
| fine-salt |
Fine salt |
3 g |
1.00% |
Fine salt. |
Method
- Combine sucrose and glucose syrup. Cook to a dark amber caramel.
- Carefully add warm cream in stages and stir smooth.
- Remove from heat and blend in butter and salt.
- Cool completely. The ripple should be thick but pourable at 20 C.
Phase Outputs
| Key |
Phase Output |
Description |
| cold-caramel-ripple |
Cold caramel ripple |
Prepared intermediate for a later Phase. Use 150 g in three 50 g layers; retain any remainder. |
PHASE D — CHURN AND LAYER
Technique Applications
| Technique |
Controls |
Purpose |
| Ice cream churning |
Draw approximately -6 C |
Churn before layering the cold ripple. |
Method
- Churn the aged base to approximately -6 C.
- Pack one-third of the ice cream into a chilled container and drizzle with 50 g ripple. Repeat twice, ending with ripple.
- Fold once with a narrow spatula to marble; do not mix completely.
- Harden at -18 C for at least 4 h.
- Retain any prepared ripple beyond the selected 150 g portion.
Caramel colour is the main flavour control. Stop before it smells burnt or acrid.
Any remaining ripple keeps refrigerated for up to 2 weeks.
Phase Outputs Used
| Phase Output |
Use |
| Aged base |
Use aged-base. Churn all. |
| Cold caramel ripple |
Use cold-caramel-ripple. Use 150 g in three 50 g layers; retain any remainder. |
Results
Carlo retrospectively reported that the recipes as recorded were all successful
and fully tested, and approved this whole formulation for production. No further
Recipe-specific observations were supplied. Cooking dates, trial counts, detailed
tasting notes and instrument measurements were not supplied. Source yields are
estimates, not measured results of this retrospective record. Recorded alternatives
have no separate variation-specific testing observations in this report.
Decision
Carlo accepted the formulation as recorded, including the explicit reviewed
clarifications, for production in the complete 25-Recipe batch on 2026-10-02.
Proceed to normal exact-version Promotion after valid serialization, applicable
Curation and reference resolution. Technical validation does not perform that
human acceptance. A materially changed formulation needs its own exact-version
evidence and acceptance; this report does not automatically follow future changes.