Daniel sent us this one, and it's a practical engineering problem disguised as a food question. He's looking to build a homemade meal replacement system — something nutritionally balanced, low-fat for medical reasons, and capable of replacing missed lunches during busy workdays. He wants to know whether powder or bars make more sense as a foundation, what the trade-offs are in terms of nutrition, shelf life, cost, and batch prep. Then, if powder wins out, he's asking for a first-principles walkthrough of ingredient selection, how to actually calculate the nutritional profile rather than just following a recipe, and what metrics to track for iterative improvement. So this isn't a "what should I eat" question — it's a systems design question where the user is the ingredient.
It's the right question to ask, because most people approach this backward. They find a recipe online, mix it up, drink it for three days, feel terrible, and quit. Daniel's starting with the constraints first — low fat, shelf stable, batchable, nutritionally complete — and then asking what format and ingredients satisfy those constraints. That's the engineering mindset. It's also, I should say, a question that comes up a lot in clinical contexts. When I was practicing, patients post-cholecystectomy would ask me exactly this — how do I build something I can keep at my desk that won't send me running to the bathroom twenty minutes later.
Right, because if your gallbladder's gone, your bile trickles continuously instead of dumping in response to a fatty meal. So a high-fat meal replacement bar would be exactly the wrong thing. Daniel didn't specify gallbladder, but low-fat for medical reasons points in that general direction, and the constraint is the constraint regardless of the underlying cause.
So let's start with the format decision, because that's the fork in the road. Powder versus bar. And I want to be direct here — for a low-fat, nutritionally complete meal replacement that you're making at home, powder wins on almost every axis that matters. But we should walk through why, because the reasons are instructive.
Alright, walk me through it. What makes bars so difficult for this use case?
It's a structural chemistry problem. A bar has to hold together. It has to be coherent, not crumble into granola dust in your bag. And the things that make bars cohere are almost universally high in fat. Nut butters, coconut oil, dates, chocolate chips, whole nuts — these are binding agents and texture elements, and they all bring fat. A typical homemade bar recipe using dates and almond butter as binders is going to run eight to fifteen grams of fat per two-hundred-fifty-calorie serving. That's thirty to over fifty percent of calories from fat. If you're targeting under five grams of fat per meal — which is where most low-fat medical diets land — bars are fighting against their own physics.
You'd need some kind of fat-free binding agent, which basically doesn't exist in nature. Or at least not in anything that tastes like food.
There are things like xanthan gum or guar gum, but those create a gummy, almost gel-like texture that most people find unpleasant in a bar format. You can use marshmallow or egg white as a binder, but now you're introducing moisture, which kills shelf life. Which brings us to the second problem — homemade bars go bad fast. Without commercial preservatives and controlled water activity, you're looking at two to four weeks before texture degrades and fats start to oxidize. Rancidity is a real concern. Meanwhile, a dry powder stored in an airtight container with an oxygen absorber can sit happily for twelve to eighteen months.
That's a dramatic difference. Two weeks versus a year and a half.
It's the water activity. Bars typically have a water activity around zero point six to zero point seven, which is high enough for mold and staling reactions. Dry powders are down at zero point two to zero point three. Microorganisms can't grow, and oxidation reactions slow to a crawl. The moisture in bars isn't just from the binding agents — it's also from things like dried fruit, which you'd need for palatability. Every gram of water you add to make it taste good is a gram of water that's working against your shelf life.
What about cost and batch prep time? Daniel mentioned wanting to make a month's supply at once.
This is where powder really pulls ahead. Thirty servings of powder — that's a month of lunches — can be mixed in about twenty minutes with a kitchen scale and a large bowl. You weigh each ingredient, dump it in, stir, and funnel it into a storage container. Bars for the same thirty servings require cooking or pressing, cutting, wrapping individually, and you're looking at two to three hours minimum. Cost-wise, a homemade powder runs about a dollar fifteen to a dollar thirty per four-hundred-calorie serving. Commercial meal replacements like Huel or Soylent run two-fifty to four dollars. Homemade bars come in around a dollar eighty per serving because you're paying for those binding ingredients. So powder is cheaper, faster to make, and lasts vastly longer.
There's one more advantage you mentioned that I think is underrated — serving size flexibility. A bar is a fixed unit. You eat the bar, that's the meal. With powder, you can scoop two hundred calories for a snack or six hundred for a proper meal, and it's the same container, same prep.
That's actually more important than it sounds. Hunger isn't binary. Some days you need a full meal, some days you need to bridge a gap. The powder adapts to that. The bar doesn't. Now, I should be fair — bars do have genuine advantages. No shaker bottle, no cleanup, no water source needed. You can eat a bar one-handed while driving. And there's a psychological satiety component to chewing that liquids don't provide. The act of mastication triggers satiety signaling through the vagus nerve. But for Daniel's specific constraints — low fat, shelf stable, batchable, cost-effective — powder is the clear winner.
We've settled on powder as the format. Now let's get into the real engineering: designing the powder from first principles.
This is the part I love. You start with the target profile, not the ingredients. Don't walk into a bulk food store and grab things that look healthy. Define what the finished product needs to deliver, then find ingredients that hit those numbers. For a four-hundred-calorie meal replacement with under five grams of fat — that's our constraint — you're looking at roughly thirty-five grams of protein, fifty grams of carbohydrates, ten grams of fiber, and the rest from micronutrients. That's your skeleton. Now you hang ingredients on it.
Let's start with protein. What's the best source for this application?
Soy protein isolate. It's typically ninety percent protein by weight, with less than one gram of fat per thirty-gram serving. It has a complete amino acid profile — all nine essential amino acids in reasonable proportions — and it's the most cost-effective option, usually running seven to ten dollars per pound in bulk. Pea protein is a decent alternative, but it has slightly higher fat at about two grams per thirty grams, and the texture is grittier — it doesn't dissolve as smoothly. Whey isolate is excellent nutritionally, but it's dairy-based, more expensive, and some people have digestive issues with it. For a plant-based, low-fat, smooth-mixing protein, soy isolate is hard to beat.
Oat flour is your workhorse. It's literally just rolled oats blended into a fine powder. It provides slow-digesting complex carbohydrates, contributes about two and a half grams of fat per hundred grams — which is low enough to work within our fat budget — and it gives the shake a pleasant, mildly sweet, oaty background flavor. The texture matters too. Oat flour adds body without turning the shake into paste. You can also add a small amount of dextrose or maltodextrin — maybe ten grams per serving — for quick energy and improved mixability. But the ratio matters. Too much oat flour and the shake gets thick and gritty. Too much dextrose and you're spiking blood sugar, which leads to an energy crash ninety minutes later.
What's the right ratio?
For a four-hundred-calorie serving, I'd start with about forty grams of oat flour and ten grams of dextrose. That gives you roughly forty-five grams of carbs, mostly slow-digesting, with enough quick energy to feel immediately satisfying. You can adjust from there based on how you feel. If you're crashing, reduce the dextrose and increase the oat flour. If the shake is too thick, do the opposite.
You mentioned psyllium husk powder as the gold standard.
One to two grams per serving. Psyllium is mostly soluble fiber, which means it absorbs water and forms a gel. That gel slows gastric emptying — your stomach holds onto the meal longer — which directly improves satiety. There's good research showing that viscous beverages can produce equal or greater satiety than solid foods through gastric distension and slowed nutrient delivery. The common myth is that bars are more filling because you chew them, but the stomach responds to volume and viscosity, not to how many times your jaw moved.
Psyllium is basically pure fiber with negligible fat.
The alternatives — inulin, acacia fiber — work too, but inulin in particular can cause significant gas at higher doses because gut bacteria ferment it rapidly. Psyllium is less fermentable, so it's gentler for most people. Start low, though. If you go from zero fiber to five grams of psyllium in one day, you're going to have a bad time.
Now, you mentioned some overlooked ingredients earlier — nutritional yeast, potassium bicarbonate, salt. Why do those matter?
Because most DIY meal replacement recipes are nutritionally naive. They focus on macros — protein, carbs, fat — and ignore the micronutrient profile and electrolyte balance. Nutritional yeast is a fantastic addition. When fortified, it provides a complete B-vitamin profile including B12, which is critical if you're replacing multiple meals per week with this powder. It also adds an umami, slightly cheesy flavor that rounds out the sweetness of the oat flour and dextrose. Three grams per serving is plenty.
This is one of those details that separates a thoughtful formulation from a random mix of powders. Protein powders, especially isolates, tend to be high in sodium from processing. If you're drinking two of these shakes a day, you're getting a sodium load without corresponding potassium. Potassium bicarbonate corrects that ratio — it adds potassium, which most people don't get enough of, and the bicarbonate helps buffer acidity. It's a tiny amount — maybe one gram per serving — but it matters for long-term electrolyte balance. Add a pinch of salt too, not just for electrolytes but because salt suppresses bitterness and enhances sweetness. It makes the shake taste better without adding sugar.
That's the kind of detail that someone would never think of on their first pass. Which brings us to micronutrients more broadly. How do you actually cover vitamins and minerals without sending samples to a lab?
This is where the honest answer is: you can't do it entirely with whole food powders. You cannot practically source every vitamin and mineral from spinach powder and ground flaxseed in a way that fits into a reasonable serving size. The pragmatic approach is to use a high-quality multivitamin powder as your micronutrient base, then supplement with specific whole-food powders for targeted gaps. Spinach powder for vitamin K, cocoa powder for magnesium and iron — cocoa powder is actually surprisingly mineral-rich, with about forty milligrams of magnesium per tablespoon — and the nutritional yeast for B vitamins.
You're not pretending this is a whole-food product. You're engineering a complete nutritional profile using whatever sources are most effective.
And that's actually more honest than a lot of commercial products that list fifty whole-food ingredients in microgram quantities just for label appeal. The multivitamin powder approach means you can actually calculate what you're getting. Which brings us to the tools.
Alright, walk me through the calculation workflow. Daniel asked specifically about how to estimate the nutritional profile rather than just following a recipe.
The tool I'd recommend is Cronometer. The free tier lets you create custom recipes — you enter each ingredient by weight, and it gives you a full micronutrient breakdown against recommended daily intakes. Here's how the workflow goes. You create a new custom recipe, name it something like "Daniel's Meal Powder Batch One." You add thirty grams soy protein isolate, forty grams oat flour, ten grams dextrose, five grams psyllium husk, three grams nutritional yeast, one gram potassium bicarbonate, a pinch of salt, and one serving of your chosen multivitamin powder. Cronometer sums everything — calories, macros, all the vitamins and minerals — and shows you a panel with percentage of RDI for each nutrient.
Then you look for gaps.
You look for gaps. Typically, calcium and vitamin D will be low, because those are hard to get outside of dairy. You might see iodine is low if your multivitamin doesn't include it. Maybe vitamin K is borderline. Now you can iterate — add a quarter teaspoon of calcium carbonate powder, or increase the spinach powder, or swap to a different multivitamin. The point is you're making decisions based on numbers, not intuition.
For someone who wants to go deeper than Cronometer's interface, you mentioned the USDA FoodData Central API and Google Sheets.
If you're the spreadsheet type — and Daniel works in tech, so I suspect he might be — you can build a Google Sheet with a reference table of ingredients and their nutritional values per hundred grams, pulled from the USDA FoodData Central database. Then you use VLOOKUP or XLOOKUP formulas so that when you enter a gram weight for an ingredient, the sheet automatically pulls in the calories, protein, carbs, fat, fiber, and key micronutrients. Sum the columns, compare to your targets, and you've got a living recipe document that recalculates every time you tweak a number. It's more upfront work than Cronometer, but you get complete control and you can track changes over time.
That also makes it easy to share the recipe with someone else or version it — batch one, batch two, batch three.
That's the mindset. This isn't a one-and-done recipe. It's a system that evolves. Which is exactly what Daniel's last question was about — the iterative improvement metrics.
You've got your recipe and your numbers. Now the real work begins: how do you know if it's actually working for you?
I'd track four things in a simple log. First, satiety — rate your hunger on a one-to-ten scale two hours after drinking the shake. If you're consistently at a six or higher, you need more protein or more fiber. Probably fiber, because protein is easier to overdo. Second, energy levels — rate your energy at one hour and three hours post-shake. If you get a spike and crash, your carbohydrate profile is off — too much dextrose, not enough oat flour. If you feel steadily low energy, you might need more total calories or more complex carbs.
Track any bloating, gas, or irregularity. If you're bloated, reduce the psyllium by half a gram or switch to acacia fiber. If you're constipated, increase psyllium slightly and make sure you're drinking enough water — psyllium without adequate water is like pouring concrete mix down a pipe. And fourth, cost per serving. Recalculate it every time you change an ingredient or buy from a different supplier. The whole point of DIY is that it's cheaper, but if you start adding exotic superfood powders, you can easily drift past commercial prices without noticing.
The red flags you mentioned are useful heuristics. Hungry within ninety minutes means more protein or fiber. Bloated means dial back the psyllium. Lethargic means check your carb quality and total calories.
The key discipline is: change one variable at a time. If you increase psyllium, swap dextrose for oat flour, and add a new multivitamin all in the same week, you have no idea which change caused which effect. Wait three to four days between adjustments. Your body needs time to adapt.
That's the kind of patience that most people — myself included — are terrible at. We want to fix everything at once.
Which is why most people abandon homemade meal replacements after two weeks. They try to optimize everything simultaneously, feel weird, and conclude the whole concept doesn't work. When in reality, they just changed too many variables and couldn't diagnose the problem.
To pull it all together into something actionable — if someone wants to build this system starting this week, what's the minimum viable recipe and the workflow?
The five-ingredient minimum. You need a protein powder, a carbohydrate base, a fiber source, an electrolyte balancer, and micronutrient coverage. That's it. For protein, soy isolate at thirty grams per serving. For carbs, oat flour at forty grams plus ten grams of dextrose. For fiber, psyllium husk at five grams — start there, adjust down if needed. For electrolytes, one gram of potassium bicarbonate and a pinch of salt. For micronutrients, one serving of a multivitamin powder and three grams of nutritional yeast. That's your starting point. Total fat is about three grams. Cost is about a dollar fifteen per four-hundred-calorie serving.
The batch prep workflow?
Buy ingredients in bulk. Soy protein isolate in five-to-ten-pound bags. Oat flour in twenty-five-pound sacks if you're committing to this long-term — it's dramatically cheaper per pound. Get a large food-grade container with an airtight lid, a kitchen scale that measures to the gram, and some silica gel packets and oxygen absorbers. Weigh each ingredient for thirty servings — multiply your per-serving weights by thirty — dump them all into the container, close the lid, and shake. Actually shake it. Dry mixing is more effective than stirring for large batches. Then store with the silica gel and oxygen absorber. Each morning, scoop your serving into a shaker bottle, add cold water, shake with the mixing ball, and you're done. No blender needed if your oat flour is finely ground.
That's a real point — the blender myth. People assume you need a high-powered blender to make powdered meals drinkable, but a shaker bottle with a wire mixing ball works perfectly well if the powder is fine enough and you use cold water. Cold water reduces clumping because it slows the hydration of the proteins and fibers, giving them time to disperse before they gel.
Warm water causes the psyllium and protein to hydrate instantly on contact, forming little gel capsules of dry powder inside. Cold water gives you a few seconds of suspension time to shake everything into solution. It's a small detail that makes the difference between a smooth shake and a lumpy mess.
Let me ask the uncomfortable question. How close does a DIY powder like this actually get to something like Huel or Soylent?
The honest answer is close but not identical. The commercial products have proprietary micronutrient blends, texture optimizers like sunflower lecithin for emulsification, and natural flavor systems that are genuinely hard to replicate at home. They've also been through stability testing and third-party nutritional verification. Your homemade version won't have that. But for sixty to seventy percent of your meals, it's more than adequate — especially if you're eating whole foods for the other meals and using this to fill gaps. The danger is when someone tries to go one hundred percent DIY meal replacement without any other food sources. That's where micronutrient deficiencies can creep in over months.
Because the gaps you don't see — calcium, iodine, vitamin D, maybe selenium — don't cause symptoms until they're clinically significant.
By then you've been deficient for a while. So the framing should be: this is a meal replacement for missed meals, not a total food replacement for all meals. If you're replacing five to ten lunches a week with this powder and eating normal dinners, you're in great shape. If you're trying to live on it exclusively, you need to be much more rigorous about micronutrient verification, and honestly, you should probably just buy the commercial product that's been formulated by food scientists.
That's a sensible boundary. What about future directions? Once someone has their baseline recipe dialed in, where do they go from there?
A few interesting paths. One is adding a greens powder — something like spirulina or a commercial greens blend — for phytonutrients that aren't captured in standard vitamin panels. Another is experimenting with different protein sources for variety. Hemp protein has a nice omega-three profile, though it's higher in fat. Pumpkin seed protein is interesting — good mineral content. You could also develop a savory version. Instead of oat flour and dextrose, use something like powdered vegetable broth, tomato powder, and herbs for a warm, soup-like meal replacement. Nobody talks about savory meal replacements, but they're completely viable and a nice change from sweet shakes.
A hot savory meal replacement powder is the kind of idea that sounds either brilliant or deeply wrong, and I'm not sure which.
It depends entirely on execution. But that's the fun of building your own system — you can try things that no commercial product would risk.
Alright, we've covered a lot of ground. Format decision — powder wins for low-fat, shelf-stable, cost-effective batch prep. Ingredient design — start with the target profile, build from soy isolate, oat flour, psyllium, electrolytes, and a multivitamin base. Calculation tools — Cronometer for most people, USDA data and Google Sheets for the spreadsheet-inclined. Iterative improvement — track satiety, energy, digestion, and cost, and change one variable at a time.
The overarching principle: treat this as an engineering problem, not a recipe search. Define your constraints, select ingredients that satisfy them, measure the results, and iterate. It's the same mindset you'd apply to any system design.
Which is probably why Daniel framed it the way he did. He wasn't asking for a recipe. He was asking for a design methodology.
And now: Hilbert's daily fun fact.
Hilbert: The adze — a wood-shaping tool used by ancient Egyptian shipwrights — nearly became the dominant agricultural implement of Madagascar around 400 CE when a trading vessel carrying a cargo of specialized bronze adze heads ran aground on the island's eastern coast. The local communities, who had never seen metal tools of that sophistication, spent two generations attempting to adapt the adze heads for tilling their volcanic soil, a task for which the tool's perpendicular blade was spectacularly ill-suited. The experiment was abandoned only after iron-working knowledge arrived via Southeast Asian settlers, rendering the bronze tools obsolete and the adze-as-hoe movement a brief, frustrating chapter in Malagasy agricultural history.
An entire civilization spent two generations trying to garden with shipbuilding tools.
The sunk cost fallacy operating at a civilizational scale.
This has been My Weird Prompts. If you found this useful, leave a review — it helps other people who are tired of overpriced, underperforming meal replacements find the show. We're back next week.
