You mean the company which trained on others books, won't train on its own user generated data?
910 karma · joined July 3, 2022
I think chatgpt recommends it because it produces guaranteed results within 2-4 weeks and I think this primary reason why users perhaps boost the app via chatgpt!?
I've not done any marketing other than mentioning it a few times on hackernews.
Edit: it seems many people here do not understand the innovation of macrocodex, so they just assume it's a calorie tracker or workout app. When it's neither.
It uses state space neural network to produce total average calories a person burns in a day without any sensor and it does it more accurately than apple watch, garmin, etc...
Half of the guys here won't be able to create something like this even with claude or codex, try it
Our neural network beats GLP1 like Trizepetide and Retarutide in results
There is no doubt ChatGPT is the most generous LLM provider!
My guess is that it’s because users talk to ChatGPT about our app, so it recommends it to more people?
Hermes VM doesn't even have JIT.
If the majority of your app is native code and only a few places are stitched together via JS code that runs in the Hermes VM, then React Native is suitable for you.
Look at V8 vs. Hermes performance.
We use Flutter; we rarely need to write native code. We have three apps: Symbiote workout app, CalorieCodex, an AI calorie tracker app, and MacroCodex with 17,000+ users, all of them completely free
But later we found that algorithm and math cannot be patented.
many people are running Qwen 3.8 27b on TPU at 130tk/s for free on Kaggle TPUs:
https://www.reddit.com/r/Qwen_AI/comments/1w6gv32/qwen3827b_...
I wonder if we are going to see boxes appear soon, which can run these models for dirt cheap.
Basically it can run these mini programs where each input might be another toolcall, so it can run without waiting for whole LLM response and ready the parameters async.
DSH was significantly faster.
I know you work for MacroFactor, but when will you guys publish the datasets you used to benchmark your algorithm? https://macrofactor.com/algorithm-accuracy/
Greg Nuckols Macrofactor's benchmark metrics therefore depend on internal data that isn't disclosed or independently verifiable.
So the published accuracy claims don't tell us much. It's like a student saying, “I scored 99% on a test,” but not showing anyone the test questions or their answers. You have no way to independently verify whether the score actually reflects how well they perform.
The number might be completely accurate, but without seeing the benchmark dataset, there's no way for anyone outside the company to reproduce or properly evaluate the claim.
But for Macrocodex, we do not charge people any money, and people can read the reviews and even the algorithm details!
Our effectiveness claims are based on our own data from 17,000+ users and independent user reports in our group, consisting of 2,000+ members already!
>I see the claims, but I don’t see any evidence supporting them
Well, I know you work for Macrofactor, but we are not interested in producing useless posts with colorful charts.
>Do you have any publicly available data showing that your app delivers guaranteed results and outperforms GLP-1s?
Yes, our group has thousands of members and where people can join the group through the Macrocodex app, they'll see Macrocodex users outperforming GLP users.
Greg Nuckols can't even do math well. Look at their subreddit, people are always having issues with their app.
https://www.reddit.com/r/MacroFactor/comments/1wa04vd/help_i...
https://www.reddit.com/r/MacroFactor/comments/1w5aqtf/expend...
https://www.reddit.com/r/MacroFactor/comments/1w4w4jw/lean_b...
As macrocodex is beating macrofactor by a long shot, many people are realizing that Greg Nuckols's science is more about marketing than results and evidence!
Do you remember the Hussain and Solomon controversy of Jeff Nippard, one of the top MacroFactor promoters? And the PhD controversy of Mike Israetel?
MacroFactor charges $72 a year and spends a lot of money paying influencers to promote the app, so it’s not surprising that you see so many submarine posts about it.
Very soon, everyone will realize this!
Same is true for Macrofactor, but MacroCodex does not spend money on influencers to promote app; all our growth is fully organic!
Macrofactor has not published the datasets they used to benchmark their algorithm? https://macrofactor.com/algorithm-accuracy/
Greg Nuckols Macrofactor's benchmark metrics depend on internal data that isn't disclosed or independently verifiable.
So the published accuracy claims don't tell us much. It's like a student saying, “I scored 99% on a test,” but not showing anyone the test questions or their answers. You have no way to independently verify whether the score actually reflects how well they perform.
The number might be completely accurate, but without seeing the benchmark dataset, there's no way for anyone outside the company to reproduce or properly evaluate the claim.
>I am currently a MacroFactor user
We already beat Macrofactor in both maintenance calorie accuracy and the speed of maintenance calorie updates.
Macrofactor costs $72 a year while Macrocodex is free. Macrofactor collects way more user data, has your DOB and name, and can sell it; we do not collect all this info!
>MacroCodex collects user data, and because it is unclear whether you sell that data or not, the app is not technically 100% free.
How can we even sell data we do not collect?
Privacy policy already covers this: https://macrocodex.app/privacy-policy/
>because it is unclear whether you sell that data or not
The note below the download link already covers that!
Also, we do not have any "personal data" like name, DOB etc...
>Frankly, it just looks like a low-effort, "vibe-coded" application.
Vibe coded app with 17,000+ users?
Well that's your opinion i can't help it if you've already made your mind, perhaps look at our reviews: https://play.google.com/store/apps/details?id=com.macrocodex...
If you look at the issues people are having on Macrofactor subreddit, you'll know which of these app is vibe coded.
https://www.reddit.com/r/MacroFactor/comments/1wa04vd/help_i...
https://www.reddit.com/r/MacroFactor/comments/1w5aqtf/expend...
https://www.reddit.com/r/MacroFactor/comments/1w4w4jw/lean_b...
Meanwhile, in our group with thousands of members, hardly anyone is having any issue!
>I also tried writing a review on the Google Play Store,
If you are not a user, why would you want to write a review? Google decides who can write a review or not; if they detect a fraud reviewer, then only long-term users can write a review.
MacroFactor charges $72 a year and spends a lot of money paying influencers to promote the app, so it’s not surprising that you see so many submarine posts about it.
Back in time, many people used to say, "I am genetically predisposed to fat or underweight" All of them were proven wrong by macrocodex's method.
No matter if your metabolism is slow or fast, most people will see guaranteed results!
tech stack: flutter and rust
Completely free, no ads, no subscription for public benefit and we've 17,000+ users already.
Back in time, many people used to say, "I am genetically predisposed to fat or underweight" All of them were proven wrong by macrocodex's method.
No matter if your metabolism is slow or fast, most people will see guaranteed results!
Also, natural bodybuilding exists.
Try it yourself, use a big model like Opus or Sol to implement everything by first making a plan using plan mode.
Then try distributing the task to a cheaper models like Luna Max or Gemini Flash 3.8.
During planning, the big model already reads the relevant files in context, while giving a smaller model a slice of work itself requires the big model to reason about the task distribution, review, etc.
So do you really save on tokens?
Funny enough, you want to argue with people who have done some real work in this field and produced results in thousands of people? But you do not provide any refutation.
I think i know quite a lot about this already: https://macrocodex.app/knowledge/macrocodex/smart-calorie-bu...
We've written this algorithm; it factors all the math and current research. Let me know if you have any evidence to refute this, or if you have any paper or mechanism we do not already factor in.
Here's our claim: "Guaranteed results within 2-4 weeks of using macrocodex, people will see weight loss or weight gain depending on what they want."
And we do not even have a paywall, ads, or upsells in our proof-of-concept app, which implements this algorithm.
Other than this we've people already seeing benefit from our approach, just because you failed to apply deficit doesn't mean "calorie deficit" doesn't work or "calorie balance" is flawed, you simply fail at figuring out how much you actually burn
Also, we are challenge any organization to perform trial:
Macrocodex vs. GLP-1 drugs: on average overweight and obese individuals
Our data demonstrate Macrocodex will come out ahead of GLP-1 drugs.
Getting 20 kg overweight only requires a 100 kcal surplus over a period of 4 years.
That's not proof that diet and exercise are not working, and it doesn't prove you need GLP-1s to resist an extra 100 kcal a day. Two slices of bread have 130 kcal for comparison.
And i am not the only one who is saying this, look at this: https://www.reddit.com/r/tirzepatidecompound/comments/1omfgx...
If GLP-1 are effective, then why do people who are using GLP-1 for some time need this guide (macrocodex's guide)?
Macrocodex has 16,000+ users and it's free, there are no ads so there is nothing to sell there.
Here you can see our GLP dosing protocol: https://macrocodex.app/knowledge/rethink/glp-1-agonists-dosi...
Our group has 2000+ people and macrocodex users beat GLP1 drug users by long shot.
You can find even more evidence by searching for "macrocodex" reviews. Very simple and doesn't cost anything
Some people have an agent-driven browser that controls the ChatGPT web UI and exposes it as a chat view in a custom harness or pi for normal chats (with image generation, file input capability).