Gulf shrimp head-and-shell koji garum

A chef’s shrimp garum on the Mississippi Gulf Coast

This is a modern, chef-documented shrimp seasoning made from the heads and shells left when Gulf white shrimp are peeled. Alex Perry’s formula adds koji, water and salt, then holds the mash at 60°C for two months before straining it. The name “garum” places the sauce in a broad culinary conversation about savory liquids made from animal material, but this particular recipe is not an ancient Roman formula or a documented continuation of one. It is a contemporary restaurant preparation; Perry’s first development date is not given.

The place matters as context, not as a claim of invention. Coastal Mississippi describes the Gulf and seafood industry as central to the region’s communities and food culture, and notes the area’s shrimping and processing history. The chef’s recipe belongs to that present-day coastal setting: it takes a familiar peeling by-product and makes it into a concentrated seasoning. The available account does not identify a traditional Mississippi shrimp garum, a local name for it, or a longer regional lineage, so none should be inferred from the setting.

A January 2021 article reproduces Perry’s Surf and Turf recipe and names him as the chef at Vestige in Ocean Springs. That is a dated attestation of this method, not a date of invention. The Coastal Mississippi article also presents the recipe among contemporary Gulf-coast chef recipes. These sources establish authorship, location and use more securely than they establish a production history. No household tradition, commercial production scale, or local supply chain for the condiment is described.

What the name does—and does not—tell us

“Garum” is used here for an enzyme-led seasoning from shrimp heads and shells. It is useful as a family resemblance: the result is intended to be a savory liquid condiment, strained from an animal-protein mash. It does not mean that every sauce called garum shares a formula, fermentation ecology, age, or safety record. The word alone cannot establish that this shrimp sauce tastes like Roman fish sauce or like Southeast Asian fish sauces.

Historical fish garums and modern koji preparations are not interchangeable. Ancient Mediterranean fish sauces were made from fish and salt by methods involving endogenous enzymes and microbial activity; this record’s sources do not describe the Perry recipe as a historical reconstruction. Modern cooks also use “garum” for rapid, enzyme-assisted seasonings whose substrates and conditions differ. The general Modernist Food method is a separate technique; its standard ratios and timing cannot fill gaps in Perry’s recipe.

Research on sakura shrimp sauce offers a particularly clear comparison because it also uses shrimp and koji. Funatsu and colleagues studied minced sakura shrimp with shells, rice koji or shrimp koji, 15% salt, selected microorganisms in some batches, fermentation at 30°C for 24 weeks, then heating and filtration. That is a controlled microbial fermentation study, not Perry’s mash of heads and shells held at 60°C for two months. Its temperatures, organisms, salt level and results must stay attached to that study. It demonstrates that shrimp-and-koji seasonings can be made by a different route; it does not supply a missing specification for this one.

Enzymes, warmth and the limits of what is known

Koji is the enzyme-bearing ingredient in the chef’s formula, but the recipe does not identify its grain, mold species, preparation, or supplier. In koji-based seasoning methods generally, enzymes made during mold growth can hydrolyse proteins: proteases cut large protein chains into shorter peptides, while peptidases can release smaller peptides and free amino acids. Those smaller compounds can contribute savory taste and provide material for further reactions. This is a general explanation of enzyme-assisted amino-paste methods, not a laboratory analysis of Perry’s batch.

The shrimp heads and shells supply the animal material that the enzymes act on. Processing them to a chunky paste increases contact between the ingredients and breaks the material into a more uniform mash. The source does not give a particle size, a target texture, or separate weights for heads and shells. It also does not report measurements of amino acids, protein breakdown, pH, salt in the water phase, or yield. The intended transformation is clear from the garum method; its measured extent in this preparation is not.

Warmth changes the rate at which enzymes act, but it is not a simple on/off switch. Enzymes have different activity and stability across temperatures, and their performance depends on the actual koji and mash. The chef specifies 60°C and two calendar months; no temperature log, turning schedule, enzyme assay, or endpoint test accompanies the published recipe. The method is therefore best understood as a long warm enzyme-assisted digestion. Calling it “fermentation” in the recipe source does not by itself show which microorganisms remain active or what microbial changes occur.

The Japanese sakura-shrimp study makes the distinction visible. It followed a multi-month microbial fermentation at 30°C, measured changes during processing, then heated and filtered the product. Those documented microbial and chemical results belong to that formulation. They cannot be projected onto the Perry mash, whose specified temperature is much higher and whose process has not been similarly monitored or measured.

Why these four ingredients are used

The recipe calls for one kilogram of shrimp heads and shells together. Perry obtains them while peeling Gulf white shrimp for the Surf and Turf dish. Saving both parts gives the seasoning its named crustacean base, while also making use of the material left from preparing the whole shrimp. The source does not specify whether the heads and shells are raw or cooked, how thoroughly they are rinsed, or how the amount is divided between them. Those omissions should remain visible rather than being filled with a different recipe’s practice.

The 225 grams of koji supplies the enzyme-rich component. Because the source only says “koji,” the grain and mold are unknown. It would be inaccurate to rewrite the ingredient as rice koji or name a particular Aspergillus species. Different koji substrates and cultures need not have identical enzyme profiles, so this recipe’s unspecified ingredient limits how precisely its mechanism can be described.

The chef adds 800 grams of water and 240 grams of salt. Water lets the crushed shrimp material, koji and salt mix into a mash that can be held together; salt seasons the mixture and affects its chemical and microbial environment. The recipe gives no measured salt concentration in the water phase and no validated preservation claim for this particular formula. Neither the quantity nor the word “salt” establishes a safe control on its own.

These amounts are the formula as published. The sources provide no tested scaling rule, substitute crustacean, alternate koji, or salt adjustment. Swapping shrimp for another seafood, using a different koji, or borrowing quantities from a quick-garum recipe would create a new preparation rather than a documented variation of Perry’s.

The sequence from peeling scraps to strained seasoning

First weigh the heads and shells saved from peeling the Gulf white shrimp. Perry’s published directions then process them until a chunky paste forms and transfer that paste to a large glass jar. The coarse paste keeps the ingredient identity recognizable while increasing the surfaces exposed to the koji, water and salt. No blade type, processing time, or fine-purée endpoint is prescribed.

Add the 225 grams of koji, 800 grams of water and 240 grams of salt, then mix thoroughly. The source specifies the weights but does not say whether the water is warm, whether the salt dissolves first, or whether a particular koji form is expected. It instructs the cook to press plastic wrap directly against the mash and seal the jar. This describes the chef’s covering and vessel, not a validated oxygen-control or preservation system.

The hold is two calendar months at 60°C. Retaining “two months” matters: the source does not define a fixed number of days, and there is no published schedule for stirring, opening, or sampling. It also gives no measured endpoint by which a reader could decide the mash is complete. The number is a record of Perry’s method, not a generally validated time-temperature safety process.

Finally, strain through a chinois into another vessel and refrigerate. Straining separates the liquid seasoning from the solids; it does not prove that pathogens or toxins have been controlled. The published recipe says the sauce may be refrigerated for up to a year, but that statement is not supported by process validation or measurements in the available sources. The structured method therefore records the chef’s instruction while keeping the absence of a validated shelf life explicit.

From seasoning a dumpling to misting the plate

Perry uses the garum twice in one dish. Two teaspoons go into the beef-cheek filling for dumplings, where the sauce seasons the meat from within. At the end, he puts more into an atomizer and lightly mists the dumplings. One use distributes it through a rich filling; the other places a small amount on the surface so the eater encounters it with the browned wrapper and accompanying shrimp.

The wider plate is built around Gulf white shrimp and braised beef-cheek gyoza, with a beurre citron and a Jimmy Nardello pepper emulsion. The shrimp garum ties the seafood preparation to the beef filling without replacing either main ingredient. This is the clearest documented use of the sauce: a measured seasoning inside a dumpling and a fine finishing application over the assembled dish. The source does not provide a standalone tasting note or a general dosage for other foods.

The combination is one chef’s restaurant composition, not a traditional Mississippi pairing. It shows how a garum-like sauce can bridge ingredients from land and sea, and how using the peelings can connect the whole shrimp preparation to the finished plate. For other uses, the available sources do not document a standard list of pairings, so the Surf and Turf dish remains the reliable example rather than a claim that this particular garum is universally suited to any recipe.

Reading the method with care

This is a source-specific professional recipe, not a validated home-preservation process. The published instructions describe a sealed jar of shrimp material, koji, water and salt held warm for two months, but do not supply process validation, a measured pH or water activity, a water-phase salt measurement, or a validated endpoint. FDA seafood guidance treats hazards such as botulism and pathogen growth as process- and product-specific. It is context for understanding why these missing controls matter; it does not validate this recipe.

The separate Modernist Food discussion of animal-protein amino pastes likewise says its general enzyme-assisted method is not validated as a botulism control. That caveat applies to the method discussed there, not as a test of Perry’s jar, but it reinforces the need not to convert a culinary instruction into a safety guarantee. Refrigeration after straining is the chef’s direction; neither it nor the source’s one-year storage statement establishes a safe shelf life for this batch.

Shrimp is a crustacean and a priority food allergen. The finished seasoning is made directly from shrimp heads and shells, so its identity as shrimp-derived must be retained when it is used or served. The recipe’s original use places it in beef-cheek dumplings and on a plate with shrimp; a reader should not mistake “garum” for a non-seafood ingredient.

Finally, the method does not state that the shrimp are raw or cooked, and it offers no sensory test for completion. Smell, taste, appearance and a long hold cannot substitute for process measurements. The account preserves the chef’s published quantities and sequence as documentation while stating plainly what the sources do not establish.