Yet another newbie looking for help


First off, sorry if this is the wrong category. But this newbie is looking for help.

I need some guidance. I am a 155lb 5’6" 31yr Male coming off a 70% Paleo diet. With a need to lose weight(damn cravings), develop muscle and clear mental cobwebs. I already have a plan to up my workout routine, but I wanted to do that after getting into soylent.

The Nutrient Profile Calculator on Diy.Soylent.Me is fairly straight forward, but I was stymied by what Carb/Protein/Fat Ratio to select. Considering my familiarity with Paleo, would Keto be a good choice?

Also, what Nutrient profile should I base mine on? What, if anything, should I adjust? If a few months from now I included Nootropics, would I need to adjust the profile to support them?

Thanks for any constructive assistance.


155 lb male and you need to lose weight? I guess I’m just used to having a heavier frame. I could barely even dream of getting below 190.

My suggestion - plan your soylent around your eating habits. If you’re a snacker like I am, it’s unlikely you’ll be able to easily change your habits on a dime. If you incorporate some caloric headroom into your soylent, it will help you ease out of snacking. In general, I don’t get food cravings anymore, even though by all accounts I’m shortchanging my metabolism by at least 400 Calories a day.
According to some calculators, with my weight, I should be eating close to 3000 Cal (!), but my full days soylent is just over 1600, and I love it. Complete fullness.

Try using a basic soylent recipe, but cut out some of the empty carbs. You don’t necessarily have to go full keto. My recipe is ~25:25:50, mainly due to flax seeds for omega 3 and cutting out all refined sugar. I’ve still got about 100g of carbs, 70 something if you discount fiber. But I’m steadily losing weight.


I should have expressed that, rather than lose weight, I was trying to lose fat. It is not a lot, but enough to bother me. Also, depending on who you ask, my ideal weight for my height and age would be 143 ish.

But back to my conundrum. Taking the last comment as a suggestion, my Nutrient Profile is now low carb. I used “QuidNYC’s DRI for Him: Male, 31-50” as a base for the profile. Mostly because I get the impression he has a better idea of what he is doing than I. And that he/she is looking for similar results.

I have been examining various recipes and trying to modify them to fit the profile. But nothing is satisfactory yet. I want to get this correct and complete the first time.

Has anyone yet developed a program to help with getting a recipe to match a profile? I am aware of someone working with a Genetic Algorithm program but I largely regard those as unreliable. Though the only solution I could derive involved either: intersecting polytopes or higher order vector addition. Not fun.

But lacking a technical fix I am hoping for someone to point out a workable recipe.

Thanks again for any help.


Why? They’re demonstrably sound and easy to implement. You can use gradient descent or other types of algorithms if you can model your problem space with a sufficient degree of accuracy, but a GA gives you the benefit of not having to model your problem space, simply apply constraints and test for fitness. It’s an elegant solution for a very complex problem. :smile:

The reason there isn’t a program out there isn’t because implementing a GA is hard, but because nobody wants to take the 2-3 weeks to create and refine the app when it usually just takes a day or so to create a recipe that fits any arbitrary profile.


As I state my opinion, please bear in mind I have never actually run a GA.

But from what I understand, the algorithm creates one or more variants, each with a small change. Then checks if any of the variants are improvements. It keeps one or more of the improved variants and repeats the process over and over. This seems like it would get stuck at local maximums and miss the ideal answer.


Any space search algorithm has the possibility of getting stuck at local optima. Gradient descent (and all other gradient/newton/etc methods, regardless of order) would need to be used alongside repeated random initial sampling or maybe large perturbations upon reaching the local optimum if you want it to have any chance of reaching a somewhat global optimum. You could instead try taking a massive amount of samples and interpolate to create some splines which you could use to more easily find a global optimum. There’s a lot of problems with that, though, mainly having to do with the massive dimensionality of the space, the necessity of using splines over high-degree polynomials, and the still-significant search time of an optimum over high dimensional splines.

Genetic algorithms (and more generally, evolutionary algorithms) have all sorts of nifty tricks to get around local optima. Instead of being constrained to the gradient of the surface, EA’s can sample in multiple directions, with varying step sizes. If you’re particularly concerned about local optima, maybe you could try making a few offspring with small mutations for a more standard gradient descent approach, while also making a few offspring with larger mutations to see if maybe things are better on the other side of the hill.

Crossover mutations between bifurcated populations can also be helpful. If each of two populations are converging to their local optima, perhaps combining their genomes to find some middle ground will find something even better.

Combine the above with a meta-strategy for evolving multiple populations scattered throughout the space, then crossing their best offspring (after a number of generations) with one another to create a new set of distinct populations, and you can end up with a really robust algorithm which maps out many optima.

I really just see GAs as a set of tools for approximating gradient descent and avoiding local optima. You can pick and choose which to use depending on your knowledge of the space, but as @jrowe47 mentioned, you can implement it naively and it’ll get you your global optimum, eventually.

Er, I wish I could contribute to the more on-topic discussion of energy source ratios, but I’ve been content with the standard ~60/20/20 carb-heavy profile. (I haven’t done all that much research, but) Given that people seem to be surviving on carb-heavy, carb-light, and carb-free (keto) diets, with weight loss seeming to correspond to caloric restriction and regular exercise more than the energy source ratios, I’m inclined to say it doesn’t matter much. The biggest draw to keto for me is hearing about eating chocolate, fat, and peanut butter as components of a balanced diet, but I like my oats.


This is one of my favorite explanations. It’s brief, doesn’t require knowledge of coding, and gets the point across without excessive theory. It also gives you enough foundation to search on if you want to delve deeper.


Back on topic.
Abandoning low-carb as a goal, I have modified a nutrient profile for myself, and modified the popular People Chow 3.0.1 recipe.
I would like some experienced eyes to give it a look over and call out any flaws I may have made.
AcerM’s Variant of People Chow 3.0.1