Miso-Fermented Hazelnut

Turning Nuts into Fermented "Cheeses" | Part 1 of 2

Hi,

It’s been a while, hasn’t it? I hope Christmas and the New Year treated you well, and, like me, you’re looking forward to a year of learning and exploration. As this is often a moment people reflect on their time spent and make plans for the future, I thought I’d start the year by returning to one of the techniques that marked my first steps into the unknown in fermentation, fermented nuts.

Cashew with Penicillium roqueforti

As you may know, my first adventures in fermentation took the form of beer, bread, and sauerkraut, and honestly, what more do you need in life? From there, vinegar naturally followed, as did kimchi and all things lacto. But it was turning nuts into umami, creamy ferments that marked the beginning of fermentation as play for me. Back then (many years ago) there weren’t many recipes for mouldy nut cheeses (a description of the technique I’m not a huge fan of, as the comparison between nut fermentation and dairy fermentation ends with appearance and umami, as both ferment, age, and cook quite differently from one another).

With no recipe for guidance, there was a lot of trial and error, but I feel I made good in the end, figuring out how to make a strong base that tasted umami, rich, creamy, and unmistakably still very nutty. One of my pet peeves is when a ‘replacement’ product bastardises the ingredients used to make it in order to become a poor imitation of something else. Hence my dislike of calling these fermented nut things cheeses. But honestly, it’s the closest ingredient to them. So, unless you can offer a better name for them, we’re stuck with nut cheeses or, my personal favourite, mouldy nut bois ( ← Welsh slang for boys).

Now, here comes a fair warning, to pull off the final boss version of these, you’ll need a modified wine fridge or personal cheese cave, but the basic version is also delicious and perfectly accessible with a normal home kitchen. So, I’ll cover the base technique here, then follow up with the final version next week, for the curious and adventurous.

---

When I first started making these, I was still cheffing full time. I decided to experiment because friends of mine who wanted to eat plant-based foods were reporting horror stories about the cheese alternatives available to them. After some early experimenting, I started to see promising results. Soon after, friends and colleagues began taking an interest, some even placing weekly orders and buying them quicker than I could make them.

I reached out to the food safety people here in the UK to find out whether I could turn them into an actual product to sell. Of course, they were utterly clueless about fermented foods and, after hours of wasted time in meetings, the best they could offer was “try pasteurising it”. Which, as you’ll see later in the newsletter, would completely ruin the final product. Mind you, this was at a time when hardly anyone knew what kimchi was, and kombucha was a rare sight in shops - and never raw. Things might have changed by now.

I eventually came to the conclusion that I’d rather spread the joy of this technique by teaching others to make them for themselves, and sidestep the waste of time jumping through hoops set by people who think all microbes are germs. After all, I trust this technique in your hands far more than theirs.

What was once the excitement of discovering something new (perhaps not for everyone, but it was new to me) has become something I make almost without thinking now. Like making pastry, it has gradually turned into one of those techniques that feels like coming home.

And that got me thinking about how something can feel like home even when it has no connection to childhood at all. It might surprise you to find out, but mouldy nuts were never a part of my upbringing. And yet the act of making them now carries a sense of familiarity that casts the idea of home in a new light, one that is less a rose-tinted view of childhood, and more about repetition. Returning to the same actions again and again, paying attention, noticing changes, and taking part. And I thought that would be a good feeling to share with you at this time of year.

It’s funny the thoughts you have about mouldy nut bois.
So, let’s get into the mechanics of soaking, blending, culturing, and ageing.

---

Index:

• Miso-Fermented Base Recipe

• Which Nuts Work

• Which Nuts Don’t

• The Science Behind it (the nerdy bit)

• How To Use The Paste In The Kitchen

---

Miso-Fermented Hazelnut Base for Aged “Nut Cheeses”

This recipe produces a high-inoculation, enzyme-rich hazelnut paste designed specifically as a base for mould-ripened nut cheeses. At this stage, the aim is not acidity or long preservation, but to build a smooth, stable, well-seasoned core that can later be shaped, dried, and cultured with surface moulds. But as it is, it makes a delicious smooth dip, or a thick, pliable paste to add to sauces or as binding agents in fillings.

The higher proportion of raw miso matters here. It introduces a dense microbial ecology, brings a meaningful enzyme load, and contributes salt in a way that supports early fermentation without pushing the paste into full miso territory.

Ingredients

(makes approximately 550–580 g fermented nut paste)

• 400g raw hazelnuts*, preferably fresh and unroasted

• 60g mineral water

• 110g raw, unpasteurised miso (rice, barley, or mixed-bean miso at 10% salt)

• Fine sea salt:

  • 9.0g to reach 3.5% total salinity, or

  • 14.7 g to reach 4.5% total salinity

* If you struggle to make this with hazelnuts, I recommend trying with cashews, which are more forgiving but less nutty.

---

Notes

• The final paste weight depends on water retention during blending; salt calculations are based on an average total weight of ~570g.

• Miso contributes 11g of salt at this ratio; additional salt is added to reach the target total salinity.

• Do not reduce salt below 3.5% if the paste is intended for shaping or further ageing.

---

1: Lightly toast the hazelnuts
(optional, but recommended)

Begin by lightly toasting the hazelnuts at around 160–170 °C, just until fragrant and barely coloured. The aim is aroma, not browning. Over-roasting creates oxidised fat notes that only become more pronounced with ageing, so err on the side of restraint. Allow the nuts to cool completely before continuing.

2: Soak overnight

Soak the hazelnuts in plenty of cold water for 8–12 hours. This hydrates the nuts evenly and makes it possible to achieve a smooth paste later with far less added water. Drain thoroughly once soaked.

3: Blanch and remove the skins

Blanch the soaked hazelnuts in boiling water for 2–3 minutes, then refresh immediately in cold water. Rub off the skins completely, which improves texture and flavour more than you’d expect. Skins contribute bitterness and grit, and blanching also reduces the native microbial load, giving the miso a clearer run later.

4: Drain well and warm slightly

Drain the blanched nuts well, then let them sit for five to ten minutes to steam-dry. If possible, begin blending while the nuts are still slightly warm, ideally around 35–45°C. Hazelnut fats shear more cleanly when warm, which means you’ll need less water to reach smoothness. This matters, because once water is fully emulsified into the paste, it is very difficult to remove later.

5: Blend carefully to a thick, smooth paste

Start blending the nuts with no added water at all. Only once the machine genuinely struggles should you add water, and then only in very small amounts – a teaspoon at a time – blending fully between additions. I found this is just about doable with 60g of water. If it’s too sticky and thick though, leave it to cool and add the miso at this stage, as it will loosen the mixture up a little without adding more water.

You are aiming for a paste that is cohesive and smooth, but still thick. Stop before it becomes pourable. This is a critical point: with hazelnuts, once you create a fully emulsified slurry, the water becomes bound and will not drain later. If the paste flows like batter, you have already gone too far for reliable ageing.

Alternatively you can leave the paste more coarse if you’re struggling to make it smooth, but this will result in a grainier texture in the final paste.

6: Cool completely before inoculating

Allow the blended paste to cool to below 30–35 °C before adding miso. Adding raw miso to a hot paste risks damaging both its living microbes and its enzyme activity.

Once cooled, mix in raw, unpasteurised miso at 25–30% of the nut paste weight. This is intentionally high as the saltiness of miso is one of the key things protecting the ferment from spoiling.

At this point the paste should taste clearly savoury and salty, but not aggressively so. It should still read as hazelnut first, miso second.

7: Adjust salt to the correct range

Miso already contains salt, typically around 10–12%, so you’ll need to account for this before adding any extra. Your target for this base paste is a total salinity of roughly 3.5–4.5%. Under-salting is the most common reason these projects fail later, especially once the paste is shaped and exposed to air.

As a sensory check, the paste should taste similar in saltiness to a well-seasoned pâté. Nuts are actually surprisingly good at balancing the flavour of a lot of salt. This much salt paired with an ingredient like tomatoes or carrots would ruin the flavour, but when mixed with high fat ingredients like nuts or high starch ingredients like potatoes or polenta it’s surprising how much it will take before tasting unpaletable.

8: Pack firmly and protect the surface

Pack the paste firmly into a clean jar or bowl, pressing out all air pockets as you go. Smooth the surface flat, then lightly salt it with a small pinch of sea salt. Press a circular piece of baking parchment directly onto the surface, then cover the mouth of the jar or bowl with cling film (or a jar lid with the seal removed). If using cling film, use a sharp knife and poke a hole in the top.

This step helps block oxygen and nutrient-rich nut pastes are an open invitation to unwanted moulds that can thrive given access to it. Dense packing and surface protection buy you stability while the interior fermentation establishes itself.

Have a taste at this stage too, by the way, and make notes so you have something to compare to later on.

9: Short fermentation

Ferment the packed paste at around 18 °C for 3–7 days. Begin tasting from day three onward.

This ferment won’t turn sour like a lacto ferment does. What you should notice is a reduction in raw nut notes, a rounder aroma, and a creamy flavour with a punch of savoury depth. Texture may also become slightly more cohesive as enzymes begin their work.

Once the paste tastes clearly improved and “set”, it’s time for step 10.

10: Chill to stabilise (and prepare for shaping)

Transfer the paste to the fridge for 24–48 hours. Chilling firms the fats, slows fermentation, and makes the paste easier to handle. At this point you should have a dense, smooth, savoury base that is ready to be used in sauces, stews, fillings, and dips. And you’ve managed to make the core body of what can be prepared for surface cultures in the next stage.

After this short room-temperature fermentation stage, the paste will keep for several weeks in the fridge, and have its best flavour in the first month.

Where this leaves you

At the end of this process, you should have a hazelnut paste that is stable, well-seasoned, and evenly inoculated. It is not yet a “cheese”, and it should not be aged openly in this form. The next step – shaping, rind development, and controlled humidity – deserves its own article, which is where this base comes into its own (see below).

---

Nuts that work well (and why)

Hazelnuts
Hazelnuts are the benchmark for this method. They’re high in fat, relatively low in starch, and develop deep, rounded aromas under enzymatic action. When blanched properly, they emulsify smoothly and produce a paste that holds together well through fermentation and shaping. Their flavour also stands up to miso without being swallowed by it, which matters once ageing begins.

Cashews
Cashews are extremely cooperative. They blend easily, require less added water, and produce a naturally smooth paste with good body. Their mild flavour makes them an excellent canvas for fermentation and surface cultures, though they tend to produce a gentler, sweeter result than hazelnuts. Because they’re so forgiving, they’re a good option for first attempts.

Macadamias
Macadamias are very high in fat and low in fibre, which allows them to emulsify beautifully into a dense, luxurious paste. They age well and develop a buttery depth that suits mould ripening. The main drawback is cost, and they require careful water control during blending, as they can tip into looseness quickly.

Almonds (with care)
Almonds can work, but only if blanched thoroughly and blended very fine. They contain more structural fibre than hazelnuts or cashews, which can lead to chalkiness if particle size isn’t reduced enough. When done well, they produce a clean, elegant base, but they’re less forgiving of shortcuts.

---

Nuts to avoid

Walnuts
Walnuts are rich in polyunsaturated fats that oxidise readily. During fermentation and ageing, this often shows up as bitterness or rancidity rather than depth. They also bring strong phenolic compounds that clash with miso-driven savouriness. Even when fresh, they’re unpredictable over time.

Pecans
Similar to walnuts, pecans oxidise easily and tend to develop muddy or stale flavours under enzymatic action. Their sweetness can also read oddly once savoury fermentation takes hold.

Brazil nuts
Brazil nuts are extremely high in fat but have a coarse internal structure and a flavour profile that can turn sulphurous or metallic when fermented. Texture is difficult to control, and the results are rarely improved by ageing.

Pistachios
While tempting, pistachios contain pigments and phenolics that often become dull or brown during fermentation. Their flavour is easily lost beneath miso, and the cost-to-reward ratio isn’t great.

Peanuts
Peanuts are legumes, not true nuts, and behave very differently. They ferment readily, but the flavour trajectory moves toward peanut pastes, miso-like condiments, or tempeh-like products rather than anything cheese-like. They deserve a different technique entirely.

Chestnuts
Chestnuts are avoided for structural reasons rather than flavour. Unlike most culinary nuts, they are low in fat and high in starch, which shifts this process away from slow, fat-driven enzymatic conditioning and toward sugar release and starch behaviour. There isn’t enough triglyceride present for lipolysis to contribute richness, and instead the paste tends to gel or sweeten as starches are broken down. The result moves toward porridge-like textures or overt sweetness rather than savoury integration, making chestnuts poorly suited to a method. They do, however, make very nice amazake.

A useful rule of thumb

If a nut blends smoothly with minimal water*, has stable fats that resist oxidation, and tastes good on its own when raw or lightly toasted, it’s a strong candidate.

If it’s prone to bitterness, rancidity, or structural grit, fermentation will likely exaggerate those flaws.

* Mineral water is used here because minerals matter to both enzymes and microbes, and nut pastes don’t bring much of their own. Calcium, magnesium, potassium, and other trace ions support enzyme activity and help microbes establish themselves more reliably, especially in a low-sugar system like this. Mineral water also offers a small buffering effect, smoothing early pH changes, and avoids chlorine or chloramine that can interfere with fermentation. It isn’t essential, but it removes a variable and makes the process more consistent from batch to batch.

---

What each part contributes – The nerdy bit

Below is my estimation of what’s actually going on in this technique and how it changes the ingredients over time. I haven’t had the process lab tested, so this isn’t a claim of certainty, but an informed interpretation based on what we know about miso, its microbial ecology, enzymes it carries, and, of course, nuts.

Once the miso is mixed into the hazelnut paste, enzymatic activity begins immediately. Proteases introduced with the miso start hydrolysing hazelnut proteins, breaking them down into shorter peptides and free amino acids. This has a clear sensory effect. Raw hazelnuts often carry sharp, drying, slightly bitter notes; as proteolysis progresses, those edges soften. The paste tastes rounder and more savoury, even though it hasn’t become acidic. Texture shifts too. As proteins are partially broken down, they interact differently with water and fat, and the paste often feels more cohesive and less pasty despite no water having been removed.

Alongside this, lipases act on the triglycerides that make up the bulk of hazelnut fat. As these fats are broken down, free fatty acids and glycerol are released. Free fatty acids contribute flavour directly, but just as importantly, they extend it. This is where the sense of depth and length comes from: the paste coats the mouth more fully, and the flavour lingers. This is also where care is needed. Free fatty acids are more prone to oxidation than intact fats, which is why oxygen control matters. If bitterness appears later in the process, it usually comes from oxidation rather than microbes.

Hazelnuts themselves contain very little starch, so amylases aren’t driving the process in the way they would in grain- or legume-based ferments. Having said that, residual amylases and glycosidases from the miso do act slowly on trace polysaccharides and oligosaccharides present in the nuts. The amounts of simple sugars released are tiny – you won’t taste sweetness – but they matter. They provide a small, continuous source of energy that allows microbes to remain metabolically active rather than shutting down entirely.

Those microbes are mainly salt-tolerant lactic acid bacteria carried over from the miso. In this environment they are alive, but strongly substrate-limited, so acid production is modest. The pH may drop slightly, but not enough to register as sour. Instead, this gentle acidification tends to reduce muddiness, soften bitterness, and make flavours read more clearly.

Halotolerant yeasts from the miso are also present. Their activity is low, but they contribute small amounts of volatile compounds that affect aroma rather than taste. Some miso producers deliberately encourage these yeasts during maturation because they play a role in developing miso’s characteristic smell. In the nut paste, their contribution is subtle: the aroma shifts from flat or neutral toward something more developed, without becoming alcoholic or bready.

Salt, finally, is doing a lot of work here. At this level it won’t preserve indefinitely or dominate flavour. It regulates enzyme activity, limits microbial overgrowth, and keeps the system stable while the slower processes take place. Too little salt and unwanted organisms gain ground; too much and enzymatic activity slows to the point where flavour development stalls. In the right range, salt largely disappears as a taste and functions instead as a pacing mechanism, setting the conditions for everything else to happen evenly.

Over several days, the paste doesn’t become sour or bubbly like other ferments. It becomes less raw, more savoury, more cheesy, and more internally consistent. If you taste it daily, the changes are incremental but cumulative. That gradual settling is the signal that the technique is doing what it’s meant to do – preparing the paste for what comes next.

---

Suggested Uses:

Base for filled or bound preparations
Use the paste as the savoury core in things like stuffed vegetables, dumpling fillings, savoury tarts, or vegetable terrines. Mixed with cooked mushrooms, caramelised onions, grains, or greens, it binds and seasons at the same time, adding depth without needing cheese or cream.

Sauce or dressing
Thin the paste with warm water, stock, or a splash of acid to make a sauce. It behaves similarly to tahini, but with more savoury persistence and less overt nuttiness or bitterness. Used this way, it can stand in for cheese-based or cream-based sauces without being a full replacement, but remember, it won’t melt or stretch like a cheese.

Spread or layering ingredient
Use it sparingly as a spread on toast or alongside bread, or layer it into sandwiches, flatbreads, and wraps. Treated like a soft cheese / hummus hybrid, it can bring richness and depth without needing to be perfectly smooth.

Fermented liaison(ish) for broths and vegetable jus
Broths are often finished with nut purées (almond in ajo blanco, walnut in Caucasian soups) to add body without cream. This paste works beautifully the same way. Whisk a small amount into a hot vegetable broth, mushroom stock, or reduced roasting jus just before serving. It thickens slightly and adds savouriness without turning the liquid heavy. The fermentation gives the broth a “long-cooked” quality, even if it hasn’t been reduced for hours.

Used carefully, it acts as a plant-based liaison, dispersed into a sauce rather than melted, where it can add body in the way butter or cheese would, without behaving like a pure fat.

Worked into a savoury sabayon or whipped emulsion
One place nut emulsions shine in restaurants is when they’re aerated or lightened – whipped nut creams, foams, or sabayon-style sauces. Blend the fermented paste with warm water or light stock, then emulsify with a hand blender or whisk over gentle heat to create a savoury, spoonable sauce. You can stabilise it further with egg yolk if desired, or keep it fully plant-based.

Served warm, this becomes something you can spoon over grilled vegetables, roasted roots, or steamed greens.

---

And that brings us to the end of part one. In the next newsletter I’ll move on to shaping, surface cultures, humidity control, and the process of ageing, the stage where this paste really begins to show what it’s capable of. If your only experience of mould-ripened nut cheeses has been something acrid, raw, or aggressively penicillin, I promise this is a different path entirely. When I used to make these regularly, other chefs and shop owners placed weekly orders and argued over what little fridge space I had to spare. So I’ll ask for your patience while I write the next instalment. Until then, feel free to share, restack, or comment if you found this useful, and thank you, as always, for reading.

If you’d like to read more fermentation or food essays, see my recipe index here. Or support this publication by purchasing one of our beautiful enamel pin badges here.

Sam