Need honest opinion

Hi there! I’d love your honest opinion, roast me if you want, but I really want to know what you think about my open source framework:

https://github.com/entropy-flux/TorchSystem

And the documentation:

https://entropy-flux.github.io/TorchSystem/

(https://entropy-flux.github.io/TorchSystem/)

The idea of this idea of creating event driven IA training systems, and build big and complex pipelines in a modular style, using proper programming principles.

I’m looking for feedback to help improve it, make the documentation easier to understand, and make the framework more useful for common use cases. I’d love to hear what you really think , what you like, and more importantly, what you don’t.

https://redd.it/1phiiru
@r_opensource

OpsOrch – Unified API for Incidents, Logs, Metrics, and Tickets
https://www.opsorch.com

https://redd.it/1phom73
@r_opensource

Beliarg is a dark, gamified productivity and finance management ecosystem.

This is an open source project - free to use, modify, and distribute. It has been reforged into a Full-Stack Web Application (PWA). It combines a React 19 frontend (built with Vite) with a Node.js & PostgreSQL backend to ensure your data survives even the apocalypse)). It features a unique "Hellish" aesthetic, turning daily tasks into "Chains", expenses into "Sacrifices", and habits into "Rituals". https://github.com/D371L/beliarg feel free to leave any feedback

https://redd.it/1phplp9
@r_opensource

RANDEVU - Universal Probabilistic Daily Reminder Coordination System for Anything
https://github.com/TypicalHog/randevu

https://redd.it/1phrmw4
@r_opensource

merox-erudite – MIT-licensed Astro blogging theme with newsletter, comments, analytics & AdSense built-in

I just published an open-source Astro blogging theme that’s now part of the official Astro themes directory:
https://astro.build/themes/details/merox-erudite/

It’s a fork of the excellent astro-erudite, but with a lot of the “real-world” stuff already implemented and ready to use:

Brevo/Sendinblue newsletter integration
Lazy-loaded Disqus comments
Google Analytics + Umami support
Structured data (FAQPage, HowTo, etc.)
Google AdSense ready
Enhanced homepage (experience timeline + skills showcase)

100% free and open-source under the MIT license.

GitHub: https://github.com/meroxdotdev/merox-erudite
Live example (my own blog): https://merox-erudite.vercel.app/ and https://merox.dev

https://redd.it/1pi1sib
@r_opensource

Built a tool to catch package.json/package-lock.json inconsistencies before npm ci fails

Hey everyone! I just published a new npm package that I've been working on, and I'd love to get some feedback from the community.

What it does:

The tool analyzes your package.json and package-lock.json files to detect inconsistencies before you run `npm ci`. If you've ever had `npm ci` fail because of mismatches between these files, this is designed to catch those issues early and explain exactly what's wrong.

Current features:

* Compares package.json and package-lock.json for inconsistencies
* Provides detailed warnings about what doesn't match
* Checks for Git installation in your project
* Verifies npm version compatibility with package-lock.json's version

Planned features:

* Automatic fixes for detected inconsistencies (suggestions/PRs welcome!)

Why I built this:

`npm ci` is great for reproducible builds, but the error messages when it fails aren't always clear about *why* your lock file doesn't match your package.json. I wanted something that could be run as a pre-CI check or git hook to catch these issues locally.

This also can be added to your CI/CD workflow, and prevent from deploying in case of an error.

Installation:

npm install npm-ci-guard

GitHub: [https://github.com/yaronpen/npm-ci-guard](https://github.com/yaronpen/npm-ci-guard)

I'm still early in development and would really appreciate any feedback, suggestions, or contributions. What features would make this more useful for your workflow?

https://redd.it/1pi1qvo
@r_opensource

Wrapper tool for Google Drive seamless integration into Linux

rclone4gdrive is an open-source tool for seamless, automated, and transparent two-way Google Drive backup on Linux.

rclone4gdrive eliminates the hassle of configuring and maintaining routinely cloud syncs by providing true "set-and-forget" synchronization directly from your Linux filesystem to your personal Google Drive.

GitHub: https://github.com/thisisnotgcsar/rclone4gdrive

This is a project I built in my free time, and it’s one of my first contributions to the open-source community. If you notice anything that can be improved or corrected, feel free to let me know or open a pull request. Any help you give to improve this tool also helps me grow as a developer, so your contributions are truly appreciated!

https://redd.it/1pi2dva
@r_opensource

I built a distributed key-value store in Rust (Raft + 2PC + custom storage engine)
https://github.com/whispem/minikv

https://redd.it/1pi3rtr
@r_opensource

Built a container management + logs viewer that finally feels right to me

hi everyone, i have been doing lots of self-hosting and running things off a vps, the most difficult thing i had to live with was all the time having to ssh into a server to debug things going on, read logs or restart containers.


So I built LogDeck. It's fast (handles 10k+ logs without breaking a sweat), supports multi-host management from one UI, has auth built in, streaming, log downloads, etc


Would love to have your feedback.

github.com/AmoabaKelvin/logdeck

logdeck.dev

https://redd.it/1pi59h1
@r_opensource

DataKit: your all in browser data studio is open source now

Hello all. I'm super happy to announce DataKit https://datakit.page/ is open source from today! 
https://github.com/Datakitpage/Datakit

DataKit is a browser-based data analysis platform that processes multi-gigabyte files (Parquet, CSV, JSON, etc) locally (with the help of duckdb-wasm). All processing happens in the browser - no data is sent to external servers. You can also connect to remote sources like Motherduck and Postgres with a datakit server in the middle.
I've been making this over the past couple of months on my side job and finally decided its the time to get the help of others on this. I would love to get your thoughts, see your stars and chat around it!

https://redd.it/1pi4zul
@r_opensource

Recommendation for privacy friendly open source software to create a (stolen) bike register?

Bike registers such as bikeindex (US) bikeregister (UK), bicycode (FR) or mybike (BE) prevent bike theft, increase chances of recovering stolen bikes and help to identify thieves. But they are not interoperable and custom solutions.

I wonder which open source privacy friendly solution could be used to create a similar 'open' register to be used by every country (or entrepreneur, bike theft insurance) which wants to use it. User would upload photo and denoscription (frame number, brand and model, colour etc., presumably in structured format), user could declare a bike 'stolen, and everybody (or just authorised users) could search/filter the list of stolen bikes by brand, frame number (fuzzy search) and then have an anonymous way to send a message to the owner of the stolen bike.

The solution should have a decent interface, not just a spreadsheet, and ideally not be easy to scrape/spam. And of course top protection of the private data.

Any sugggestions what would work best, and how much work would be needed to adapt it to the denoscription above?

Thanks a lot in advance for your help!

https://redd.it/1pi7ycs
@r_opensource

My Android opensource project: DayExam

A powerful Android application designed to help you efficiently parse your school paper and store in your phone, then you can study anywhere
It is simple but useful.
opensource at github, link: https://github.com/newerZGQ/day\_exam

or you can download it at fdroid: https://f-droid.org/packages/com.gorden.dayexam/

https://redd.it/1pi3n2g
@r_opensource

polluSensWeb - webhook support added

polluSensWeb is a lightweight web-based serial interface and charting tool for visualizing and logging data from UART pollution sensors (PM2.5, VOC, etc). 

No installs, no drivers — just plug it in and open the page.

As for now, by default, JSON  configuration supports the following sensors already (in the drop-down list in the web interface):

1. Panasonic SN-GCJA5
2. Honeywell HPMA115S0-XXX
3. Air Master AM7 Plus
4. Plantower PMSA003-S
5. Plantower PS3003A
6. Plantower PMS1003
7. Plantower PMS5003
8. Plantower PMS7003
9. Plantower PMS6003
10. Plantower PMS9103
11. Plantower PMS3003
12. Nova PM SDS011
13. Sensirion SPS30
14. SHUYI SY210
15. TERA NextPM
16. SenseAir S8 004-0-0053
17. SenseAir S88 Residential
18. SenseAir S88 LP
19. SenseAir S88 GH
20. SenseAir K30
21. SenseAir K33
22. SenseAir eSENSE
23. SenseAir S8 004-0-0017
24. SenseAir K33 ICB
25. Sensirion SCD30
26. More coming soon...

PolluSensWeb just gained a powerful new feature - HTTP webhook support.

The app can now push every parsed sensor frame directly to any endpoint you choose, using customizable headers and JSON body templates.

The coolest part: both headers and body support placeholders (e.g., {{field:PM2_5}}, {{ts}}, or full field loops), letting you map sensor data into any API format without touching the code. This makes it dead-simple to forward PM readings into home automation systems, databases, online dashboards, or your own custom server.

Webhook requests can be triggered on every packet or at a user-defined interval, and a built-in “Test Send” button helps verify output instantly.

https://redd.it/1pibh6e
@r_opensource

Passless — a Virtual FIDO2 / Passkey device and client for Linux
https://github.com/pando85/passless

https://redd.it/1pibsmz
@r_opensource

I decided to open source my social platform project

Over the last few months I have been working on a project that allows you to spin up your own self-hosted social platform, complete with a web interface, iOS/Android client apps, and a robust backend.

https://axionnode.com

It is similar to Reddit in terms of functionality (communities, posts, user accounts, comments, upvote/downvote), except now with it being open source, you can customize it for your personal or corporate use, and then run your own version of a social platform.

The frontend is React Native using Expo, and you can build clients for iOS and Android easily. On the backend it uses Django / Python.

I originally designed it to work seamlessly with Google Cloud Platform (using Cloud Build, Cloud Run, Cloud SQL, etc), but I’m hoping that others may assist me to make it cloud agnostic, and add improved video functionality support.

Thanks for checking it out. Fork away and let me know if anyone might want to contribute to it.

It has been a labour of love and a fun project to try and tackle.


https://redd.it/1pidz0z
@r_opensource

Rant I'm completing my first serious project but looking back it mostly feels a waste of time

I love technology and programming but as I'm approaching the release of my first "grown-up" open source software (a software needed by school in my local community and that probably will be adopted by many other school in my region since they all share that niche need) I wonder if open source programming is a worthy investment of my limited time.

I totally believe in the beauty of having open source software implemented with love (especially in this age of enshittification where even a simple app to split expenses is ad-filled to the brim) and in the importance of digital sovereignty the issue is... people around me (and I'm pretty sure around many of you) don't care about this nerd stuff and its totally okay but at the same time its very hard to stay motivated when people close to you perceives you as a loser who spends many nights each week staring at funny code or an idiot which could "make bank with apps" but wastes his time giving away his work for free.

The other big motivations which pushed me to embark in open source programming were the opportunity to upskill and improve at day job and the sheer fun in building something without the constraints I have at my 9-5 programming job but I'm gradually finding out that in jobs once you get your foot in the door "playing the game" and selling yourself is much more important than actual skills and while I had definitely many fun and creative moments writing my application I'm not sure they're worth the expenditure of mental energy they costed. Even surfing Reddit is fun but unlike programming it doesn't require significant effort so I may as well do that or... use that time and energy to do volunteering that actually benefit people around me in more immediate ways than "free custom school software", both makes much more sense from an utilitarian POV.

Said that even if at the moment I'm pretty demotivated what I'm planning to do is to stay disciplined, complete the project and give it the maintenance and bugfixes it needs (it's not a complex software so I don't expect many bugs), regardless if its going to be fun or unfun. I'm still grateful that I was trusted to do this project and I want to repay the trust with a good job.

I'm just wondering if it makes sense to keep programming as an hobby, I enjoy it and already had many other projects and stuff to learn in the pipeline but considering the negligible job benefits and "negative" social benefits maybe its better to invest that time in:

\- Stuff I still enjoy but takes less effort
\- Stuff which gives me more tangible benefits
\- Stuff which gives other people tangible benefits

https://redd.it/1pif5kv
@r_opensource

I built stay-active - keeps Microsoft Teams showing "Active" on macOS

Problem: Teams marks you "Away" after 5 minutes. No setting to change it.

Solution: A shell noscript that simulates natural activity (mouse + keyboard) at random intervals.

GitHub: https://github.com/sleekhost/stay-active

Tech: Bash + cliclick

Install: One curl command

Size: \~6KB

Would love feedback!

https://redd.it/1pibqzu
@r_opensource

Snapchat now charges for >5GB Memories — so I made a free open-source downloader that actually works

Snapchat now wants you to pay once your Memories exceed 5 GB, and their official export tool is unreliable — some files download, some don’t, and it still shows “100%” even when large parts are missing.
I built an open-source downloader that fixes this by parsing the memories_history.html, reliably fetching every memory, correcting timestamps, adding EXIF metadata, extracting overlays, retrying failed items, and cleaning duplicates.
If your Snapchat export is incomplete or inconsistent, this solves the problem properly.

Repo:
https://github.com/ManuelPuchner/snapchat-memories-downloader

https://redd.it/1pig90p
@r_opensource

Here’s a project I made to facilitate my researcher life
https://youtu.be/_Y-NQ0-TpJ4

https://redd.it/1pih8gx
@r_opensource

Building a new way to reason with LLMs (we're also paying contributors to the repo)

Training reasoning models is really expensive and I had a suspicion that there was a lot of performance to be gained by exploring the models states better.

I’ve open-sourced a lightweight framework for latent-space reasoning, and the results have been more interesting than expected. With no fine-tuning and no access to logits, it consistently outperforms baseline outputs across a range of tasks just by evolving the model’s internal hidden state before decoding (including being able to solve problems that the base model struggles with). This uses a minimally trained judge (200 samples on a simple scorer; cost less than 50 cents to do completely) and preexisting models with no other tuning.

It works with any HF model, and the entire pipeline is intentionally simple so people can tear it apart, extend it, or replace pieces with better ideas. I’m putting up bounties for improvements because the goal here isn’t to claim we’ve solved reasoning, but to build a shared playground for exploring it. We're already collaborating with researchers in 2 of the top 5 AI Labs in the world to extend this with more sophisticated mechanisms (especially around aggregation and projections) but would love to have you guys in as well.


Let's make sure the new generation of reasoning is open source--

https://github.com/dl1683/Latent-Space-Reasoning

https://redd.it/1pimeg1
@r_opensource

How to get started with open source as a new CS grad?

Hey what's up y'all. I just graduated with a undergrad in CS and have been working as a software engineer at a mature tech company for about 6 months. I've learned quite a lot about how large scale applications and services are built and engineered, and I'm very appreciative of it.

However I'm soon going to a different company (better pay + standby flight benefits) where I'll work as a data engineer, but the actual engineering is much weaker there, and the projects I work on will be smaller scale and internal. I'll also be more accountable for my own work so I won't really have much senior help in engineering and designing of solutions.

But I still want to become a better software engineer overall as I see myself eventually going back into big tech/AI or quant (I'm doing a masters degree in ML, have undergrad degrees in applied math and CS).

I think the best way to hone my skills at that point is to become an open source contributer to well maintained projects, but I honestly don't know where to start. Just picking up issues, or reading forums all seems so daunting and hard to even begin.

For starters, my biggest problem is understanding large codebases. At my current job, I eventually understood mine better due to extensive architecture notes and just working on stuff for 40 hours a week. Obviously I wont have that same time or support level in open source software. GPT makes it easier to get started and reason about a codebase, but past that, it's still hard to work on software I'm not familiar with at all, my current job is my first experience with that, and its about to end :(


Second is the long term motivation. I think my job is very interesting, and the product I'm working on applies the concepts I learned in college very well, but ultimately I'm still doing it for the salary. I have a lot of hobbies outside of work, and staying motivated to stick to a project long term, for free, may be an issue. I dont know if that means this type of work just isn't for me, but I'd appreciate tips on how to actually stay committed to this stuff for no extrinsic reward.

https://redd.it/1pimwc8
@r_opensource