frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Show HN: TownSquare, a tiny presence layer for websites

https://townsquare.cauenapier.com/
105•cauenapier•16h ago•39 comments

Show HN: StartupWiki – A Free Alternative to Crunchbase

https://startupwiki.tech/
171•shpran•12h ago•57 comments

Show HN: Make PDFs look scanned (CLI or in the browser via WASM)

https://github.com/overflowy/make-look-scanned
98•overflowy•9h ago•48 comments

Show HN: We post-trained a model that pen tests instead of refusing

https://www.argusred.com/cli
77•dk189•14h ago•37 comments

Show HN: My Windows XP portfolio with working Game Boy and iPod

https://mitchivin.com/
56•mitchivin•8h ago•30 comments

Show HN: Microcrad – Micrograd Reimplemented in C

https://github.com/oraziorillo/microcrad
66•oraziorillo•3d ago•21 comments

Show HN: Tiny – An interpeted dynamic langauge with inline Go native functions

https://github.com/confh/Tiny
38•confis•9h ago•6 comments

Show HN: Ember, a native iOS Hacker News reader I built around accessibility

https://github.com/DatanoiseTV/ember-hackernews
90•sylwester•11h ago•21 comments

Show HN: Rocannon – Any Ansible module to MCP Tool, record sessions as playbooks

https://github.com/msradam/rocannon
2•msradam•3h ago•0 comments

Show HN: Domphy – plain-object UI for tool apps, AI-correctable output

6•khanhhuunguyen•2d ago•2 comments

Show HN: Agentic coding workflows built on Git worktrees and task evidence

https://github.com/alex-reysa/glueRun-go
5•alexreysa•4h ago•0 comments

Show HN: Criterion Closet as a website – pull any of 1,247 films off the shelf

https://the-criterion-closet.vercel.app
21•olievans•14h ago•7 comments

Show HN: Codeflowmap – map a codebase's read/write/auth data flows

https://github.com/man-consult/code-mapper
3•brian-m•4h ago•0 comments

Show HN: FERNme – agent memory that updates with ~zero LLM calls

https://github.com/mirkofr/FERNme
3•mirkofr•4h ago•0 comments

Show HN: HN Game Stories – mini-documentary of games that hit the front page

https://video.intellios.ai
2•coolwulf•4h ago•0 comments

Show HN: Vitrus – the company brain that tells you what it doesn't know

https://github.com/ahmetvural79/Vitrus
3•ahvural•4h ago•0 comments

Show HN: Talos – Open-source WASM interpreter for Lean

https://github.com/cajal-technologies/talos
101•mfornet•2d ago•27 comments

Show HN: Metiq: a real time 3D globe for 100 public datasets

https://metiq.space
143•rakeda•4d ago•42 comments

Show HN: TLA+ Process Studio

https://tlaplus-process-studio.com/?example=meeting-lifecycle
6•uptodatenews•5h ago•0 comments

Show HN: Onsight: pay a local to photograph any spot in San Francisco, on demand

https://onsight.photo/san-francisco
7•pro_methe5•5h ago•2 comments

Show HN: Notedog – Git-friendly portable Markdown journal, edit from a laptop

https://notedog.run/
4•hibariya•5h ago•0 comments

Show HN: GitGauge – A Way to Try Tell If a Repo Is Authentic

https://github.com/Davey2Waveyy/gitgauge
4•davey2wavey•6h ago•1 comments

Show HN: Gerrymandle - Daily puzzle game where you redraw electoral districts

https://gerrymandle.cc/
227•realmofthemad•2d ago•77 comments

Show HN: Lean bulk, cut, body recomp. Calculate maintenance calories

https://macrocodex.app/
22•faangguyindia•2h ago•5 comments

Show HN: LoopFlow – loop engineering for Claude Code

https://github.com/faisalishfaq2005/loopflow
3•faisalishfaq•7h ago•1 comments

Show HN: MiniPCs.zip (charting the pareto frontier of MiniPCs)

https://minipcs.zip
3•yathern•8h ago•0 comments

Show HN: Persona.js – a vanilla-JS agent UI library with native WebMCP (MIT)

https://www.persona-chat.dev/
10•becomevocal•8h ago•12 comments

Show HN: Are You in the Weights?

https://www.intheweights.com/
458•turtlesoup•2d ago•243 comments

Show HN: An ASCII 3D Rendering Engine

https://glyphcss.com
9•apresmoi•9h ago•5 comments

Show HN: An n8n alternative where coding agents build the workflows, not humans

https://velane.sh/
4•abhishekraj272•9h ago•0 comments
Open in hackernews

Show HN: Lean bulk, cut, body recomp. Calculate maintenance calories

https://macrocodex.app/
22•faangguyindia•2h ago
A very simple idea: when you eat more than your maintenance calories, you gain weight; when you eat less than your maintenance calories, you lose weight.

By using an algorithm, we can accurately figure out your maintenance calories more accurately than traditional regression based formulas like katch mc ardle.

It's way more accurate than calorie burn tracking devices like fitness bands and watches. (garmin/apple watch/whoop etc...)

Traditionally, people often use static TDEE calculators which often over or underestimate for some by 100s calories.

Chatgpt and TDEE calculators like Calculator.net or TDEECalculator.net use the same formulas, so they share the same limitation

If a beginner asks ChatGPT, "What are my maintenance calories?", ChatGPT can give them a number. But ask how it arrived at that number, and it will usually explain that it used a formula like Katch McArdle, Harris Benedict, or Mifflin St Jeor to calculate BMR, then layered activity on top using an activity factor, PAL, or MET tables.

Dig deeper and those formulas come from statistical regressions based on averages from past populations. That means maintenance calories calculated this way can be off by hundreds of calories.

MacroCodex uses your calorie intake and weight data to figure out maintenance calories specific to your body, not the population average.

It usually reaches good accuracy after about 3 weeks, or 21 days of calorie and weight logging.

This app is completely free, no paywall, no subscription and no ads. (works offline)

Most people start seeing weight gain or loss within 5 weeks.

We've reached 13,000+ users. Full support is provided if you experience any issue, most issues are resolved under 24 hours.

Comments

p1024k•2h ago
This is very helpful to me.
pipi3066•2h ago
It’s a great product. I’d mainly like to know: how does this compare to connecting Apple Health data to Claude or the ChatGPT app and having them perform calculations in a specific way?
faangguyindia•1h ago
Macrocodex can work even when calorie logging is sparse. A user may fail to log data for a few days each week, and maintenance calories can still remain accurate because it operates over your entire history.

If you read our privacy policy, you’ll see that it collects total calorie and weight data from users, which we use to improve our adaptive TDEE algorithm. It does not collect email, phone number, date of birth or anything which is not necessary for product improvement/diagnosis.

You can probably achieve something similar with Claude and a bunch of scripts if you log your calories every day.

macrocodex is entirely deterministic, it just works and users don't need to fiddle with it.

Log your calorie each day, log your weight weekly and it will continue producing calorie and macro targets for you to achieve desired rate of weight gain or loss.

It also has many more features, such as lean bulk, cut, and body recomposition modules. It includes a planner that helps you decide when to switch from cutting to lean bulking (or vice versa), or when body recomposition efficiency starts dropping for you.

For example,

To gain least fat while gainng muscles, Lean bulk is what you need.

The key is to focus on your rate of weight gain, not a specific calorie number or surplus percentage.

Weekly Weight Gain Rate:

Beginners: 0.25-0.5% of body weight

Intermediates: 0.25-0.4%

Advanced Lifters: 0.1-0.25%

Usually, a calorie Surplus: +5-10% of TDEE (roughly 100-300 kcal/day) may get you there but if it doesn't adjust your calories up or down to reach the desired weekly weight gain rate.

This is where macrocodex can do better than chatgpt/claude over long horizon.

pipi3066•47m ago
That sounds great, but it feels inappropriate to collect users' calorie data without explicitly informing them.
faangguyindia•39m ago
Calorie and weight data collected by the Android app looks like this:

`datetime | 2500 kcal | 90 kg`

By using Macrocodex, users become part of a shared data pool. They pay nothing, but they can choose to share this data.

We've this data from 17,000+ users that's what makes this accurate.

Data collection on Android can be disabled by turning off Diagnostics under Profile.

The app does not collect anything beyond this, so there is no way to link the data back to any specific person. It does not even ask for your name.

Furthermore, users can easily delete all collected data by going to *Profile > Clear Data*. During onboarding, the app will also show the Health Connect permission request.

On the web app, there is no reliable way to store persistent data because IndexedDB storage can be wiped by the OS or browser. Because of this, it requires backend to save this data.

There are no cookies used on site, no tracking pixels etc...