projects to make me employable

I am currently still in school but think ahead a little, and since uni will probably consume a lot of time on its own I plan on building a bunch of neat little projects to put on my github that would basically show a potential employer that I know programming. I am especially interested in data science and statistics, security and network stuff. My current skill level is mediocre, I know python and C, at a pretty good level with the former and a beginner's level with the latter. I've also recently been getting into CTFs and have toyed with a bunch of libraries and technologies over the last years, but never had any idea good enough to put in a lot of dedication to build any larger project. The major reason for this is that I don't really find any motivation for doing something that has been done before in a better way, and most ideas I came across have already been built. I intend to major in physics and it would be very neat to find a job or at least a paid internship that would be at least somewhat related to the kind of work I might be getting after I have my degree (that is something related to the fields I mentioned before hopefully, I don't think I will ever be smart enough to get into a research position). Pic related mostly just contains rather boring or useless stuff, at least in my opinion, maybe you can get something out of it.

I wouldn't recommend building "a bunch of neat little projects". That stuff is just gonna gather dust, a Indian will come along, replace the commiter name, and show up at the jobinterview claiming it's his. Are there no opensource projects that you use from time to time? Why not submit a PR there? Often there's lots of little annoyances left unaddressed, or even if you think it's already perfect, there's a big list of open issues. Working in other people's code is much more educational than playing with yourself, and when the interviewer asks "what's project x" you'll be able to say "my fix helped thousands of users"

>>23064 >Are there no opensource projects that you use from time to time? Until the projects he uses has a lot of dependencies, may use make file or whatever making it almost impossible for a new user(of the project) to compile their code. fuarrrk gimp.

>>23065 Not only the compiling and dependencies are the problem, most big projects have a group of core contributors and only outsource typo correction etc to new contributors. Bedt bet would be to find sec vulnerabilities, they can't reject that

>>23062 Just work on whatever you want to. Internships are more important for employability than projects unless you make something truly special (think big and modern). If you want a physics related job you should familiarize yourself with matlab and you'll probably need to get at least a masters, so look for undergraduate research opportunities to put on your grad school application. >>23065 >>23067 Why so much doomsaying about open source? If you use Linux even casually you should at least be able to follow build instructions. The only time I've ever had trouble building a project was cross compiling Ardour for Windows because fuarrrk paying for free software. I think they left it poorly documented on purpose. Build systems are complicated because they solve a messy problem. It's much less tedious to use one than to manually maintain different build systems for different platforms. There are only three types of projects that won't merge your code if you solve a problem semi-competently: massive bureaucracies, one man shows, and dead projects. If you submit useful patches to a small or medium sized project that you use, I almost guarantee they'll be accepted. I have a handful of patches in an embedded operating system that I briefly played with among other things.

>>23062 Contribute to the openzfs kernel

>>23071 >not that good with C >just work on a kernel lmao

>>23076 Some of the best code around to learn from. It has to be stable and solid. If you stick to the core components for learning, then you can become quite learned.

>>23077 But I doubt that they would accept OP's PR or even consider telling him how to do things right

Alright so I just looked through >>10408 but there was not a single project that inspired me to build anything. Many cool ideas but I don't want to replicate them.

Maybe some cool AI stuff, you have to be creative

>>23091 Have you got anything specific in mind?

>>23120 Not that lain, but here's a few ideas: >GPT for chan text >GPT for twitter text (and then send it after blue checkmarks) >Image recognition trained on hentai (categorize) >Image generator for caricatures >Image generator for hentai >Text translation from lazy typing to proper english >Text translation for obfuscating writer style They're all dumb, but you might find something interesting to play with.

>>23120 game AI that isn't retarded

>>23126 Thank you! A couple of notes on the ideas: >GPT for chan text >Image generator for hentai >Image recognition trained on hentai (categorize) Please keep in mind that I want something to put onto my CV kek >Text translation from lazy typing to proper english >Text translation for obfuscating writer style Very interesting ideas, have you got any ideas for the training data for the second one? >>23126 >>Image generator for caricatures Probably fun but not very useful and I would probably just end up stealing someone elses code as similar things have been done countless times. >>23135 I like it, any cool games in mind? Supertuxkart would probably take a little too much processing power. Could anyone please tell me how much this kind of AI training would cost? My crappy thinkpad won't survive any of it so I have to go for muh cloud training. I guess Text translation would be the cheapest? Are there any suited cloud computing services that are not run by google or amazon?

>>23139 If you don't mind selling your soul to Google, then Google Collab is a good resource for beginners. Lots of examples, and "free" (paid for by your data) time on machines that will definitely be more powerful than your laptop.

>>23139 >Please keep in mind that I want something to put onto my CV kek I hereby double down. You definitely need to do a hentai AI. Use https://github.com/psal/anonymouth to generate source texts. Also, you can prolly create a datamining thread where you tell people to "obfuscate text" (provide a source, like a news article, and then rewrite it). It'd be like a community project. Everybody pitches in data, and you give them a model. >Game AI Starcraft 1 is apparently quite accessible. You can tinker with Screeps also (JS or your favorite transpiled language). You could also tinker with Doom or Quake. >Training I don't have hard numbers since it depends on your model size and training data. That being said, a VM with a cloud provider with GPU would likely do pretty good for a reasonable price.

Find projects that can make you money. Find business needs and do them better than the competition. Then you won't need a job :) Unfortunately almost all of that stuff is in JS and Electron. If you want a fancy C project, write your own windowing system to replace xorg or contribute to Wayland since they are pretty bad. If you do a good job that will get you enough publicity someone will want to hire you. Do the stuff no one else wants to do but needs to be done

>>23160 >Unfortunately almost all of that stuff is in JS and Electron. This unfortunately. I would rather make something cool that doesn't get me money than to do some webdev soulless codemonkey work that i have zero fascination for. I assume that I am too bad at C to work on those projects and I don't really want to switch to wayland anyways, Xorg works fine for me. But thank you anyways, I appreciate every idea and this would be the kind of stuff I am interested in, I guess I am just too inexperienced for it. I had a good idea for a project I could do, basically adding a small feature to a program I use quite often but the code is hardly documented and I am not sure if I will be able to implement it. Not mentioning any specifics because I don't want to link my github account to this thread. One other thing I would like to learn about is procedural generation and generative art, so if anyone has some project ideas or learning resources for that it would be amazing.

>>23162 I've been wanting to get into that field for a long time now so i can share some resources I came across, take it with a grain of salt though because I never made any programs like that myself. https://invidious.site/channel/UCmtyQOKKmrMVaKuRXz02jbQ very impressive physics simulations and especially worldbuilding and gamedev soykaf https://invidious.site/watch?v=LrEvoKI07Ww You have perhaps seen this video, very nice ideas of small little projects https://invidious.site/channel/UCrv269YwJzuZL3dH5PCgxUw fractals, especially 3d ones with rendering and also some other cool graphics and geometry stuff maybe also coding train, he did some interesting ones the reasons I din't yet make anything myself are >I have soykafty hardware incapable of rendering complex stuff >I hate js or c# and the bloated as fuarrrk libraries that are usually used for generative art, maybe I should learn sfml though. Please let me know if there are any other good technologies to do that kind of things >I'm not the best with math

rolling. Ill do it in C if it's possible... wish me luck lainons !

>>23254 SFML is a pleasure to work with. I would recommend making a simple raycasting engine with it to begin. You can do it in C, which is also very fun. Raycasting is a very well documented technology and you can easily find tutorials on how to get started !!

>>23263 Not the person you're replying to but should I use C or C++, assuming I know neither?

>>23264 Most exemples are in CPP, so I assume it would be quicker to get started with. C is simpler overall though. This means that your code will be easier to write, read and comprehend, that is, of you have no prior programming experience

>>23262 Is it just me or are the difficulty assessments on the chart all off? >>23162 >>23254 >generative art Check out Inigo Quilez's articles: https://www.iquilezles.org/ He's one of the founders of shadertoy which is a great site for learning how to use fragment shaders for generative art. Of course you can do the same thing locally by yourself with a shader viewer, but you can't browse other people's work as easily.

>>23268 Thanks, where would you start with that? He has many articles and while raymarching seems to be most interesting I don't think I should start out with that.

>>23062 >HTML + Javascript debugger What is that supposed to be?

>>23269 Don't be intimidated to start with raymarching. The worst of it is setting up all the rays in perspective. If you still just want to dip your toes in, 2D signed distance fields can make cool patterns without raymarching. They use the same concept.

>>23271 I just looked at it, read a couple of blogposts, watched a couple of videos, but I think I can't do this. My math knowledge is too little to make it possible for me to write my own raymarching renderer or my own fractals I think. Guess I'll just stick to 2d stuff for now, perlin noise stuff looks like a good start I guess

>>23062 Learn math, a lot of math. Then learn math again. Programming is nowadays skill that even morons can learn. I would even dare to say that morons are biggest group of people who learn programming.

>>23311 That's one of the reasons why I want to study physics, besides the fact that it is fascinating I think the math can be helpful in finding a non-codemonkey programmer job. I currently read How To Prove It to learn some math that is not taught in school. But I usually have more motivation to learn a concept in math if I know of some application, so gnerative programming and shader stuff as it was recommended in this thread is probably a good project to start with. But please feel free to give further advice or project ideas, I'm happy about any input I can get

>>23312 >further advice or project ideas modelling physical phenomena is a good place to start because the algorithms have all been developed but you'll gain exposure and understanding by implementing them. pick something that ties into your other interests. >music write a string synth. there are well known differential equations that model the behavior of a plucked string. you'll learn about them, and also about numerical methods for solving them (look up runge-kutta). >physics lots of options here. beyond just doing the kind of newtonian kinematics stuff you use in game engines, there's also lots of fun to be had with simulating fluid dynamics or orbital mechanics. >crypto, pseudorandom number generators, hash functions, etc. good for learning algebra and number theory. besides traditional crypto stuff there's also applications in computer graphics. look up perlin/simplex/worley noise. >macrobiology, pathology, neuroscience, machine learning, economics, game theory might seem strange to lump these all together, but they tend to use similar models. basically you have a lattice of interconnected nodes each with its own local state. then you look into the dynamics of the thing as a whole. will teach you differential/difference equations and a bit of dynamical systems theory.

>>23325 Thank you so much, this is perfect inspiration. Orbit mechanics is my favorite idea, when I start doing the simulations in 2d it should be reasonably easy but I can make harder versions over time. I would be very happy if you could go a little more into depth about the following topics: >crypto Obviously this is a great idea and shows important applications of math, but I don't really know what kind of project I could work on besides reading a book. Have you (or anyone else) got experience with the cryptopals crypto challenges? >macrobiology, pathology, neuroscience differential equations Interesting idea but could you please share some examples? The only thing I could find was the Lotkq-Volterra equation but I don't really know what to do with it, although the wikipedua article was a fun read

>>23343 >I don't really know what kind of project I could work on besides reading a book You could implement crypto standards for learning purposes of course. But if you want something you can actually use then hash functions have plenty of applications outside of the high stakes world of security/encryption. The noise algorithms I mentioned are a good starting point. Maybe try making a terrain generator for a game engine or write a liquid shader. >Interesting idea but could you please share some examples A relatively simple one: https://en.wikipedia.org/wiki/Hebbian_theory >Lotkq-Volterra equation but I don't really know what to do with it Write a simulation based on it. Your program will take values for those four parameters, and initial predator/prey numbers, then model what happens to those numbers over time. Can you find numbers that lead to equilibrium? What about periodic oscillations between predator and prey dominance? Can you figure out how to generalize the model to a deeper "food chain"? What about modelling more complex interactions between multiple species?

>>23162 >>23254 That is so impressive, are any lains here active members of the demoscene and want to share how to get into GLSL shaders? I found the book of shaders and started with it, any motivational links and resources would be greatly appreciated

>>23377 There are no demoparties in coronatime

>>23312 Physics? OK so calculsu and linear algebra with quaternions (they are important in 3D simulations for avoiding huge rounding errors) for doing simulations and physics sim engines are obvious foundations. But be prepared that univ/college level ways of taching this stuff absolutelly is like kicking dead whales down the beach monkeys' cocks especially if you go there without any prior knowledge of differentials and/or integrals. You would be also adviced by me to pay at least some more than basic attention to probability and statistics. That stuff is basis for monte carlo simulations important in physics and for so called "Machine Learning" that is a glam name for applied statistics done on computers. The next branch that would be beneficial for you after learning calculus, algebra and statistics is going to be Finite Element Method. After that you would be able to write any physic sim you want and any 3D engine you want all with realistic mechanics and kinematics. 3D editors and CAD/CAM software also are included here. You would be able to process signals, images, sound. It's hard to think about any area of physcis related math that you will not be able to do by yourself after such math background as at this level you posses most common applied math aparatus. I would like to note that you may be eventually interested in more symbolic side of math - like foundations of mathematics in shape of first order logic and set theory plus some topology. They are typically taught as a part of computer science program at top tier univs during 1st year of undergrad program as separate classes. You kinda learn them anyway if you learn univ level algebra and calculus but not in same extent. If you go trough all that mentioned stuff you will be able to learn pretty much any area of math you want at any point of your life if needed. As of programming - try C and assembly - it's a must needed to not be yet another illiterate code monkey. Also try functional programming languages and concept. And do not buy all that hype about OOP being super duper bright and clever way of programming.

>>23391 >If you go trough all that mentioned stuff you will be able to learn pretty much any area of math you want at any point of your life if needed. Well, yes, but to get near the research fronteirs takes at least a few years of dedicated studying

>>23392 >Well, yes, but to get near the research fronteirs takes at least a few years of dedicated studying Depends on field. There are many that require from you just fluency in math aparatus to understand basic papers introducing you to the field. Rest is just systematic everyday work. Many people have tendency to overestimate amount of material that is really required to start.

>>23391 You wouldn't download a bra

>>23391 >>23392 >>23395 >>23414 But it's the new brazier frontier

Make a Bitcoin wallet

Cryprography. Proves you know math and programming very well.

>>23645 >Cryprography. Proves you know math and programming very well. Or are capable of translating algorithm diagram or psuedocode into language of choice. Which is not yet "knowing math and/or programming very well". It's just entry level.

>>23647 That is true for every project that has been done before though

>>23648 yep

>>23649 >>23648 Any advice for how to come up with novel ideas to prove I can work on projects that haven't been done before?

>>23651 >Any advice for how to come up with novel ideas to prove I can work on projects that haven't been done before? If you mean "not yet implemented as a part of real life projects" I would reccommend looking into research papers. "Carmarck's way" so to speak. He basically implemented BSP related papers done by some guy from academia and gathered a lot of fame while barely anyone even remembers name of a researcher who developed algorithms and data structures Carmack made us of. Bitcoin is also an example of implementing research paper. And original paper predates bitcoin by like a decade and a half. But that would probably require that you didn't sleep too much during math classes at your 1st year at Univ if you studied computer science. It's a lot easier at least to read and process mentally these papers in wide enough range if you are familiar with such level of math.

>>23070 > Internships are more important for employability than projects unless you make something truly special yeah but how will I get an internship without any projects? it doesn't matter if I have a 4.0 GPA in my no-name university when I'm competing against Ive League kids who go to hackathons and have a dozen projects. It feels like even the internship market is twice as competitive as it was a few years ago

>>23725 Ivy*

>>23311 I'm starting to think that people with Masters and PhDs are going to be more sought after in the following years, especially as emerging buzzwords morph into actually cohesive fields/specializations. Does anyone here with experience feel the same?

>>23064 >I wouldn't recommend building "a bunch of neat little projects". That stuff is just gonna gather dust, a Indian will come along, replace the commiter name, and show up at the jobinterview claiming it's his. This is insane paranoia.

>>23727 >Does anyone here with experience feel the same? No, it's probably the opposite. AI used to be the domain of researchers, but now it's common to see it done by enterprise programmers everywhere. Programming used to be the domain of electrical engineers and mathematicians, but now you can program without knowing those fields. Electrical engineering used to be the domain of physicists, but now you can build circuits without having to know about hamiltonians and particle physics. In every field, the bar has been lowered.

>>23733 >but now you can build circuits without having to know about hamiltonians and particle physics This has always been possible, given that the understanding of what even is particle physics came way after many electrical circuits have been built.

>>23727 I think fluctuation in demand is met by the current system. Most PhDs are funded by research money, so it's simply a fact that trendy topics will have a larger supply of graduates. Obviously there's a scarcity of grads at the start and a surplus at the end of a fad, but with something like AI we're in the middle. >>23733 >In every field, the bar has been lowered. I strongly disagree with this sentiment. It's just another form of nostalgia for the mythical old days where things were simpler and programmers were real programmers. Everything is more complex now, so it's only natural that the experts are more specialized which comes off as less capable to people looking in. >>23658 Implementing papers is a great exercise. One of my friends got most his job offers thanks to being able to talk in depth about an area he read a ton of papers from in his senior year (they were only tangentially related to the work he does now)

>>23738 >I strongly disagree with this sentiment. It's just another form of nostalgia for the mythical old days where things were simpler and programmers were real programmers. Everything is more complex now, so it's only natural that the experts are more specialized which comes off as less capable to people looking in. Sorry, I was too dismissive. A modern electrical engineer definitely has advanced specialized knowledge that would take a physicist as much time to learn as it would take the engineer to learn the physicists knowledge. (Assuming similar jobs. Give or take.) I was trying to say that the success of a field leads to more turn key solutions, more industry positions, and a lower barrier of entry for industry positions. PhDs will always be in demand, but within their field they should expect to be increasingly outnumbered in industry over time. What was once one field's PhD dissertation becomes a more specialized field's undergraduate course. This is not a value judgement on the quality of new fields. It's a recognition that new fields are created because there aren't enough PhD candidates to meet industry demand.

I'll be locking this thread or merging it with the following: >>23831

>>23064 >That stuff is just gonna gather dust, a Indian will come along, replace the commiter name, and show up at the jobinterview claiming it's his. >what's signed commits have fun trying to change the name of a PGP signed message

>>23835 Why though? This is a project idea thread, most of the stuff ITT isn't related to employment at all

>>23391 HOT O T

>>23835 why

I made a PoC botnet with P2P c2 that mimes monero. It has some interesting features I wanted to try and I'm overall pretty proud of my code. I never used this for malicious intent obviously but now I am considering if it is a good idea to publish the source code to use in a portfolio. It's decent code but it could be used for illegal activities, do employers care about this?

>>24686 I suppose the thing to do would be to check if it trips any anti-viruses and then try and dress it up as something non-malwarey. If it trips anti-viruses (it probably will, most monero miners do now) then you might as well release and then pretend its actually designed to allow quick hassle free installation of a monero miner on a network you control.

rolling

>>24686 Post your soykaf code you limp wristed kuck. The only way the malware could be tied back to you is _if_ you pussyfoot around making the digital trail start from a public source that could've been forked and run literally anywhere rather than keeping the evidence in secret code words that literally only exist on _your_ blue smoking gun of a central spook processor smdh

>>23263 Eventually after SFML, it'll be best to learn proper SDL2.

>>23062 >44. Fizzbuzz (BONUS: In Assembly) global _start section .data fizzmsg db "Fizz", 0x0a buzzmsg db "Buzz", 0x0a len equ 5 section .text _start: mov ebx, 1 push 0 fb_loop: pop eax inc eax cmp eax, 101 jge exit xor edx, edx mov ecx, 3 push eax div ecx mov eax, 4 cmp edx, 0 mov edx, len jne buzz jmp fizz exit: mov eax, 1 xor ebx, ebx int 0x80 buzz: mov ecx, buzzmsg int 0x80 jmp fb_loop fizz: mov ecx, fizzmsg int 0x80 jmp fb_loop boom

>>25446 Isn't fizzbuzz supposed to print fizz if your number is divisible by 3, buzz if it is divisible by 5, fizzbuzz if it is divisible by 15, and the number otherwise? Why does your program only print fizz if the number is divisible by 3 else it prints buzz?

>>25444 Unless you're using some of the addon libraries (image, ttf, etc.), SDL isn't worth it. Input, window management, and sound are a few hundred LOC per OS. Ultimately nothing in the grand scheme of things, and you learn more about program design and the platforms by porting it manually.

>>23062 okay lets roll ps is 4chanx working here?

>>23062 roll (unless it's hard difficulty or above - I got my own projects for that)

>>25474 > Basic neural network > "Basic" what did he mean by this?

>>25475 Single layer perceptron with sigmoid activation? I love how ReLu seems so retarded compared to sigmoids, but just works.

grind leetcode and lie in your resume

employment thread >>23831

>>25990 >goes into a thread made in 2020 to bitch about why it's not written as a post in a thread from 2021 what is wrong with you

>>25995 yet you had to bump the thread.

>>25996 impossible when every single post i made on this website is saged

Any tips on choosing a final year project? I don't really have any ideas in mind other than a machine learning system that detects a user's handwriting and replicates it. Not sure how applicable or good this is though.

>>23062 Maybe make a toy version of some product made bya company you are interested in? Also rolling rolling rolling

`>>23120 Here's something I told my friend. He's a gamer/tabletop player, so I got the games around that > Tetris learns to play using Neural Network/some other meme AI thing: 1 month to learn to code, 2 to learn to AI > Fire Emblem AI learns to play with another neural network/other meme AI (different from tetris, plus use some chess programming theory: 1 month to learn to cheat game, 2 months for AI programming probably (over estimate I feel) > Create stardew valley bots that can evaluate what needs to be done to get girl/grow crops at expected conditions and soykaf and basically plays the game itself. 1 month of learning how game works memory wise, 3-5 months for making the bot logic itself If you want to get started use guidedhacking to learn how to hack games.

A hex editor is a decent project. Not as trivial as it might seem to handle large files correctly, design a useful UI, read/write process memory, etc. Plus, there are lots of specific useful features that you could implement and actually use. Converting data to code, parsing file headers, stuff like that. An example: https://github.com/WerWolv/ImHex

Rolllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllin'

Rolling for a hopefully challenging and fulfilling project

Not strictly a "project to make me employable", more like a "fun project related to the topics I like", but the thread gradually transitioned into that so whatever. Does anyone have any ideas about what kind of project one may make using group theory? The only idea I can come up with so far is a reimplementation of the Group Explorer webapp (https://nathancarter.github.io/group-explorer/index.html) but I'm not sure if it'll be challenging enough. Any advice would be incredibly appreciated

Being employable is more about marketing yourself anyway, just focus on the things you want to learn about. For data science there is plenty of stuff out there to play with If you want to get into research your best approach is trying to network with people already there, not really focusing on having a cool github page. It's not about being smart is about being able to see the opportunities before someone else takes them.

roll, I need to make a kewl project for uni

>>23062 Rolling for something interesting

>>29694 Oh god, oh fuarrrk ;_;

>>29694 search for 'raytracing in 1 weekend', it's much more do-able than you think it is

Python or learning Rust Dice roller said Rust Rollin for project

>>29755 Web page crawler in rust? fuarrrk no

>>29756 Should be pretty straightforward in Rust. Why don't you want to do it?

>13 Music streaming (Difficult) Is music streaming really that difficult? Hm. Maybe I'll have to do some research on it... I was thinking of doing something like this for myself so I can stream my collection of music from a NAS. Getting tired of lugging it around on an SD card. (especially now that my phone has trouble with it and keep disconnecting it from my phone for no reason.)

Finally got a job opportunity after a whole year of being a NEET. It's just web-dev, and unfortunately it's a fullstack position. As much as I prefer a cozy server-side job, I don't mind. I can tell I'm going to have a good time finally making money again, especially since it's in something I poured countless of hours into just to gitgud at.

I'm in a similar position, just read through the whole thread and still don't know what to do with my free time now. I envy people with creativity and ideas so much.

>>24010 Just remove the PGP signature, git filter-branch does that implicitly. Employers won't be looking for them in the first place.

>>24010 >>30436 And even if the employers look for a signature, they can just add their own signature on.

I'm going to apply to some internships. Something for summer. Probably am too late by three weeks and all the good stuff is taken but I'm fine if it's a simple junior job for a CV. I just don't want to have another gap like during Covid.

>>23732 Insane paranoia it may be, but that is reality. Be sure to plant a 'failsafe' in your projects. Something that leaves the thief afraid to steal again.

rolling for project... aaaaaaaaaaaaaaaaaaa is the body long enough now?

Rolling. I'll see what I can do, some of these seem more like toy ideas. Also good on you for getting into CTFs OP, they are pretty fun and there's a lot of tools you can write to do them better. Including for reverse engineering.

rolling i always end up never finishing my projects anyways

>>23062 I'm rolling for terry! I will make a kernel no matter my numbo.

I'm content with making myself a bootloader, it's been on my list anyways. Rolling

i just want to rollll forever and ever

>>23732 The Indian forking your project and claiming it as his in some way is not the problem - that's basically how open source is supposed to work. If I fork someone's repo, build it and play around with it, maybe fiddle with code for some feature, it's not outrageous for me to say "I've worked on a project X that does Y". The actual problem is that many people interviewing have no idea how to tell the difference between a confident faker and someone who can actually create projects. Not that it matters where I work but for me its blindingly obvious when someone throws some garbage up on Github just to say it's there. At least it shows that they know how to use Git and have some level of ambition. Point is, if you're trying to get a job, it's the appearance that matters more than function in most cases. Having nice badges and a good readme is more important than how 1337 your code is. If someone cares enough to actually ask you about your project, that's a positive sign for you about the company -- the interview goes both ways.

I wonder what is OP doing right now and also rolinng

>>23064 >I wouldn't recommend building "a bunch of neat little projects". That stuff is just gonna gather dust, a Indian will come along, replace the commiter name, and show up at the jobinterview claiming it's his. For job searching, why not do this yourself? Take a smallish project, understand how it works on a high level, than remix/refactor it and claim it's your own. Hell if you wanna be legit, you can just fork it, add a feature/two, and then talk aobut how you added that feature. >>23070 I'm with this anon, projects are valuable only early in school, especially if you're new and have no idea how to program something with depth. I would recommend going to your CS department, and finding a professor who would let you research with them. >>23085 The nature of software projects is you can't do stuff just for the sake of it, without it being a time sink. I can noodle on a guitar and make a love song to Rei for fun. Doing something similar for programming would be boring AND more time consuming. As such, I recommend you find topics you're interested in, and make a project in that, even if it recreates something you like. For me, I like music production, and one thing I like to do if I'm stuck is use my RNG script to help inspire me. The script is ```python kicks=["Kxx", "KxK", "KxK", "xKx", "xKx"] snares=["Sxx", "SxS", "SxS","xSx", "xSx"] hats=["Hxx", "HxH", "HxH", "HxO","xHx", "xHx"] meld=["O...", "O.O.", "O.OO", "OOO.", ".O..", "..O."] pendatic_scale = list(range(1,5)) from random import choice def rando_lst(length, *lst): for i in range(length): for meme in lst: print(choice(meme), end="\t") print() rando_lst(4, pendatic_scale) ``` at some point, I can get midi to my clipboard, and then I can instantly check if the pattern sounds good. That's just an example, but even something simple like "Diablo Build Planner" would be fun and interesting for a resume, since it shows passion.

>>33794 Probably. It sounds quite difficult to me. Good on you if you know how. By the way, your roll landed on ray tracer, neat coincidence.

>32 - Chan aggregator Pretty interested at trying my hands at this. I bounce between alot of chans myself. Anyone have material I can use to study this topic?

>>35143 look at kurobaex body too short body too short body too short

https://github.com/codecrafters-io/build-your-own-x This repository is a compilation of well-written, step-by-step guides for re-creating our favorite technologies from scratch. https://github.com/vicky002/1000_Projects Mega List of practical projects that any one can solve in any programming language!

>>23062 i will try rolling :(

>>23062 at this rate I'm not sure I will ever be employable people WITH DEGREES are struggling

Bumping because I saw a (now non-existent) thread similar to this thread and made me think of it. >>35709 Thx for the suggestion! I'll take a look at it. I forgot I made that post here.

>>35711 You could also take a look at https://github.com/mayfrost/guides/blob/master/ALTERNATIVES.md and implement your alternative to something

>>23062 le rolle, will give it a good attempt I swear !

>>23064 insane >>23062 I think it's a good idea. Doing a bunch of projects is how I got my first job in tech despite only having a Physics degree, and I was explicitly told that my personal projects were what made me stand out. Many of my interviews revolved around discussing these projects, and they were extremely useful to have on hand to talk about. After you get your first "real" job they won't be relevant anymore, so make sure you are OK with that. If you do not enjoy making personal projects then it probably isn't worth the effort, but if you do enjoy it then I would go for it. I would focus on making many (~5-10) small projects rather than 1 large one, that way you can be exposed to many ideas and can honestly say in interviews that you have worked with X concept or Y language. Things to try: - Screwing around with APIs - Analysing large amounts of data (eg I used the Lichess database of 60M chess games) for one project - Picking a paper and implementing it (eg Boids is a classic)

>>23391 I'm learning programming with functional programming languages from the get go, do you think it's a good move to no be yet another illiterate code monkey?

I should really try to fix on one language and learn it thoroughly instead of learning some from many. Also roll