DEV Community: hardik singh

I created a tool that helps you learn from YouTube faster and more efficiently!

hardik singh — Thu, 03 Apr 2025 10:05:51 +0000

This week, I officially launched my tool, Kozu! 🎉

Kozu is designed to help you cover YouTube videos faster, more efficiently, and with better organization. With Kozu, you can add videos you want to watch and set deadlines to complete them, ensuring you stay on track with your learning. The intuitive dashboard provides a clear overview of all your courses, making it easy to manage your progress.

Additionally, the Tasks page gives you real-time updates on whether you're keeping up with your schedule or falling behind, helping you stay accountable. The goal is to make learning through YouTube more structured and effective.

Check it out and let me know what you think! 🚀 Here are some screenshots:

Once again, do check it out https://mykozu.xyz/dashboard
Thank you!

QuizMe.mp3: Turn Your Audio Lessons into Fun Quizzes

hardik singh — Mon, 25 Nov 2024 07:34:21 +0000

This is a submission for the AssemblyAI Challenge : No More Monkey Business.*

What I Built

Quizme.mp3 is a platform where students and learners can upload their audio lessons and receive a generated quiz with 10 questions based on the content of the audio file. Users can then attempt the quiz and receive a report of their responses along with an accuracy score. The website is built using HTML, Tailwind CSS, and the Flask web framework. It integrates AssemblyAI's Lemur for audio transcription and dynamic question generation.

Demo

Visit Quizme.mp3

Demo Video

Journey

To create quiz questions from uploaded audio files, I used LeMUR with AssemblyAI's Lemur model. First, the audio file is transcribed using AssemblyAI’s Transcriber. Once we have the transcript, I craft a prompt that asks Lemur to generate 10 multiple-choice questions based solely on the transcript’s content. The questions are designed to be clear, accurate, and directly related to the audio, with the options and correct answers formatted in a neat JSON structure. The Claude 3.5 Sonnet model of Lemur is then used to generate these questions, and the result is returned to the user. This approach ensures the quiz is tailored to the specific audio content, making the whole process dynamic and context-driven.

realhardik18 / quizme.mp3

CuriShot: A Minimalist File Sharing and Organization Tool for Students

hardik singh — Sun, 13 Oct 2024 15:37:05 +0000

This is a submission for the The Pinata Challenge

What I Built

CuriShot is a simple, minimalistic yet functional web app that allows college students to save code files and screenshots from software like MATLAB, LTspice, etc., during their lab classes. These files can be stored in the cloud and organized with tags, making it easier for students to access them later from their home systems. As a college fresher, I wanted to create something minimalistic yet intuitive, which led to the development of CuriShot.

Demo

CuriShot is live on this domain. register for an account and get started!

https://curishot.onrender.com/

My Code

realhardik18 / CuriShot

a webapp to organize and store code and screenshots take during my college lab classes

CuriShot: A Minimalist File Sharing and Organization Tool for Students

This is a submission for the The Pinata Challenge

What I Built

Demo

CuriShot is live on this domain. register for an account and get started! https://curishot.onrender.com/

My Code

{% embed https://github.com/realhardik18/CuriShot %}

More Details

Pinata was used to store and retrieve all the files uploaded by users onto CuriShot. The IPFS hash of each uploaded file was saved in the user's profile on our MongoDB database…

View on GitHub

More Details

Pinata was used to store and retrieve all the files uploaded by users onto CuriShot. The IPFS hash of each uploaded file was saved in the user's profile on our MongoDB database. Along with Pinata, the web app was built using Flask, Tailwind CSS, HTML, CSS, JavaScript, and MongoDB.

How YOU can type faster

hardik singh — Mon, 02 Jan 2023 08:11:16 +0000

Typing is a very underrated skill. Typing faster can not only make you seem cool but will also get your work done quicker. In this short article, I'll tell you how you can get a typing speed of about 100WPM in about 90 days!

Understanding the keyboard

The first step to achieving this goal is to understand and get familiar with the layout of your keyboard. to get started study, your keyboard and slowly try to type short sentences without looking at your keyboard itself. Don't worry about speed at this point. The purpose of this is to get familiar with your tool.

Now all you got to do is just practice!

Once you can type shorter sentences without looking at the keyboard, All you have to do is work on doing this faster and faster. You can track how quickly you have been typing by taking typing tests on the internet. One of the most amazing websites to do this is monkey type. You ideally want to take the 30 seconds tests which are perfect for this stage. even typing for 15 minutes daily consistently can improve your speed weekly.

Resources for typing faster

Now ill be listing some resources which are FREE to use. These can help you reach your goal of typing faster!
1)https://www.keybr.com/
2)https://monkeytype.com/
That's pretty much it, These two resources can help you reach your goal. It's now all just up to you for practicing on these websites!

Introducing BookWheat, The ultimate twitter bookmark tool

hardik singh — Sun, 28 Aug 2022 18:50:02 +0000

Overview of My Submission

i always had an issue of saving important resources and tweets on twitter. so to tackle this issue i built bookwheat. bookwheat is a twitter bot which can save the tweets which you feel are important. alongside this it also supports discord integration! this means after a quick set up, you can now automatically send the tweets which you feel are important straight to the discord text channel of your choice!

Submission Category:

Wacky Wildcards

Video Explainer of My Project!

you can understand how bookwheat was made and the which tools were used, by watching this YouTube video!

Language Used

BookWheat was made using python!

Link to Code

realhardik18 / Book-Wheat

a twitter bot which can categorically save your tweets and supports discord integration!

BookWheat, The ultimate twitter bookmark tool

BookWheat is a Twitter bot that helps you save important tweets which are relevant to you categorically! it also supports discord integration, which means with a quick setup, you can now send important tweets to your favorite discord text channel!

Overview video!

Here's a short video that explains how BookWheat was made, and the tools it uses such as Redis,Tweepy,Heroku and Discord.py

#todo

How it works

How the data is stored:

the data is stored in the form of redis keys. each key stores the information related to a particular user the file db_methods.py has various methods to retrive the data and store it in a particular key, the bot stores the twitter handle and user id of the associated user, along side this it also stores the data of all the saved tweets of that particular user this data has the name of…

View on GitHub

Check out Redis OM, client libraries for working with Redis as a multi-model database.
Use RedisInsight to visualize your data in Redis.
Sign up for a free Redis database.

the dog api!

hardik singh — Thu, 04 Aug 2022 12:38:51 +0000

i really wanted to share this api called the dogs api which is really easy to use and you can build very cool projects with this!

overview

this api is the largest collection of open source dog pictures. all these images were taken from the stanford dogs data set
you can also submit pictures of your own dogs and add it to the data set

endpoints

there are a handful of endpoints available on this api

list of all breeds available
random image of a dog
get images of dog with a specific breed
get sub-breeds of a breed here is a more in-depth overview of the enpoints

have fun with this api! :)

learn to make a RESTful API with flask!

hardik singh — Sat, 30 Jul 2022 08:29:00 +0000

so what are we building here?

we will be building a very simple and basic API which can showcase some stats of various countries with a pre existing dataset in the form of a csv. to begin, download this dataset and save it in your working directory

getting the data from the csv

first we will get the data from the csv and format it all in the form of a dictionary, so that we can display this dictionary via our api

#this bit is to set up the csv file as pandas data frame object. it will now become easier to work with our data!
import pandas
dataset = pandas.read_csv('countries of the world.csv')

now we will write a function which can give us the data of a specified country!

def InfoAboutCountry(country):
    headers = list(dataset.head(0))
    country_data = dataset.loc[dataset['Country'] == country+' '].values[0]
    return dict(zip(headers, country_data))

so first we make a variable called 'headers' which will store all the parameters in the form of a list.
now we retrieve the data of the specified country using the loc function.
finally we zip them together and then explicitly convert it into a dictionary!
we will get the following output if we call this function for 'Australia'

now on to making the api!

we first import our modules and then make a flask app and then initialize our api with the flask app which we just made

from flask import Flask
from flask_restful import Resource, Api, reqparse

app = Flask(__name__)
api = Api(app)

now we make an api instance and add a Resource to it. Then we make a parser object which can help us to parse multiple arguments in the form of a single request. and finally we add an argument which needs to be parsed(in our case country)

class CountryAPI(Resource):
    def get(self):
        parser = reqparse.RequestParser()
        parser.add_argument('country', type=str)

we then get the argument and call our function and return the data

class CountryAPI(Resource):
    def get(self):
        parser = reqparse.RequestParser()
        parser.add_argument('country', type=str)
        country = dict(parser.parse_args())['country']
        data = InfoAboutCountry(country)
        return data

now to finsh things up we add this resource to our api and run the flask app

api.add_resource(CountryAPI, '/data', endpoint='data')
app.run()

final output

as you can see our api returns the data we require!

code

you can find the code here!
https://gist.github.com/realhardik18/6d9975e8b21989afadd836c800df8b31

best resources for practicing data structures and algorithms

hardik singh — Mon, 25 Jul 2022 10:57:06 +0000

today ill be sharing my favorite sites and places to practice dsa for FREE!

why even need dsa?

practicing dsa will help you improve the way you write code and can be differentiating factor between a good developer and a great developer. alongside getting you 6 figure jobs, it will also help to optimize your code and make it more readable!

on to the resources!

project euler contains a series of challenging mathematical and programming language. this requires a balanced need of both math and programming. the problems range in difficulty from easy to hard and are aimed at people who are keen to practice their problem solving skills
codeforces has various contests and problem sets. you can solve problems and get rated on the basis of how you perform in these competitions. it also has a large and growing community of around 600K people
codewars.com is one of my personal favorites. over here you level up based on Kyu/Dan Ranks. This rank system has borrowed from a system in Japanese martial arts. one can level up by completing katas(problem sets) which are at or above their level. harder the kata, the faster you level up
boot.dev has a collection of 17 tracks on python, JavaScript and go. it includes an environment where the user can learn the required concepts and then immediately implement in a code playground integrated into the website. it also makes it really easy to track your progress and keeps you engaged as it has a ton of badges and achievements to collect as you code!
-hacerrank
has kits and problem sets based around questions which are to be asked to asked in job interviews. companies like Amplify, Quora, and Capital One use hackerrank in their technical recruiting process

conclusion

if you want to master dsa, all it takes is practice. so just pick a couple of places and consistently practice problems! if you have any other resources, feel free to share them in the comments below!!

everything you need to know about carbon!

hardik singh — Sun, 24 Jul 2022 09:55:39 +0000

yesterday, on 23rd July carbon-lang launched a YouTube video with cpp north. in this post ill talk about everything you need to know about carbon, the experimental successor of c++

why do we need carbon in the first place?

c++ is the dominant programming language for software which focus on performance. it also has massive and constantly growing codebases and investments. unfortunately, it is struggling to further improve. it is hard to improve it because of years of accumulating technical debt and challenges with its evolution process.

plus points of carbon

it's performance level(which is really important) matches that of c++. c++ is still one of the fastest languages today
existing c++ libraries can easily adopt carbon
carbon has a gentle learning curve(plus points if you are familiar with c++) and it's syntax is similar to that of c++

carbon can be built on top of c++'s existing ecosystem. this means the already existing investments, codebases and the most important aspect, developers will also be part of the new carbon ecosystem

as the developers of carbon themselves said that "Carbon aims to fill an analogous role for C++"

goals of carbon-lang

the main aims of carbon are to support performance critical software, code which is easy to understand, write and read, fast and scalable development. there is a detailed goals document which you can check out too!

project status

carbon currently is still an experimental language and it will take a time before we can see it being used in the mainstream market. the developers are trying to better understand weather they can build a language which meets their successor language criteria and also gather a critical mass of interest within the c++ community. they are currently focused on getting more feedback and participation from the c++ community. here is the document which lists their objectives and goals for the year

here is the carbon-lang repo on github

carbon-language / carbon-lang

Carbon Language's main repository: documents, design, implementation, and related tools. (NOTE: Carbon Language is experimental; see README)

Carbon Language:
An experimental successor to C++

Why? | Goals | Getting started | Join us

See our announcement video from CppNorth

Fast and works with C++

Performance matching C++ using LLVM, with low-level access to bits and addresses
Interoperate with your existing C++ code, from inheritance to templates
Fast and scalable builds that work with your existing C++ build systems

Modern and evolving

Solid language foundations that are easy to learn, especially if you have used C++
Easy, tool-based upgrades between Carbon versions
Safer fundamentals, and an incremental path towards a memory-safe subset

Welcoming open-source community

Clear goals and priorities with robust governance
Community that works to be welcoming, inclusive, and friendly
Batteries-included approach: compiler, libraries, docs, tools, package manager, and more

Why build Carbon?

C++ remains the dominant programming language for performance-critical software with massive and growing codebases and investments. However, it is struggling to improve and meet…

View on GitHub

LinkedIn API Python Library

hardik singh — Sat, 23 Jul 2022 08:19:00 +0000

LinkedIn API Python Library

here is what we will be covering in this article:

why get data from LinkedIn with python in the first place?
introduction & advantages
download and setup
retrieving profile information
retrieving company details
remove connections
get conversation details
disadvantages and possible alternatives
conclusion & project ideas

why get data from LinkedIn with python in the first place?

well im glad you asked! getting data with python from LinkedIn has many benefits. first of all, it's faster. REALLY FAST compared to a human being. let's say you're an HR manager at a company and you have to retrieve information from about 100+ profiles and create a database. this would take forever if a person were to do this. and on top of this, there is a huge scope of human error. there can always be a mistype or an invalid entry which to some extent might ruin the whole database. that's why automating this process will both save you time and save you from potential errors!

introduction & advantages

we will be using the linkedin-api python library. we already have some advantages of using this library.

it just needs the credentials to your LinkedIn account to work
it is very easy to set up and has a variety of endpoints which we will discuss further in this article # download and setup to download this library, we can simply run this pip command in our terminal pip install linkedin-api now we will create a creds.py file in our project to store our login information!

#use the information associated with your linkedin account
email = '********@gmail.com'
password = '**********'

now we can make a main.py file. we will also be making all of our functions in this file

from linkedin_api import Linkedin
from creds import password, email

api = Linkedin(email, password)

over here we first import the password and email which we saved in the creds.py file and then use that information to log in to LinkedIn!

retrieving profile information

before we start retrieving data from this endpoint, we need to get the public id of the LinkedIn account whose information we want to scrape. this can be done by getting the value between the last 2 slashes in the url of the profile. for example in the case of 'https://www.linkedin.com/in/hardik-singh-0b9ab4225/', the public id would be 'hardik-singh-0b9ab4225' we can now use this to get some data.

info = api.get_profile('hardik-singh-0b9ab4225')
print(info)

this gives us a dictionary that contains useful information like country of origin, birthday, first & last name, url to the profile picture, education, skills, and work experience
the whole list of available keys in this dictionary is given below:

dict_keys(['lastName', 'student', 'geoCountryUrn', 
'geoLocationBackfilled', 'entityUrn', 'headline', 
'industryName', 'locationName', 'geoCountryName',
 'elt', 'profilePictureOriginalImage', 'birthDate', 
 'industryUrn', 'firstName', 'profilePicture', 
 'geoLocation', 'geoLocationName', 'location', 
 'backgroundPicture', 'backgroundPictureOriginalImage', 'displayPictureUrl',
  'profile_id', 'experience', 'skills', 'education'])

you can further get the specific info by getting the value of these dict keys. for example, to retrieve the education of the user, we can do this:

print(info['education'])

#output:
[{'entityUrn': 'urn:li:fs_education:(ACoAADiU3o0BpIMjtuPu3LEwZgu6XmN8_zdNY3s,785025469)', 'school': {'objectUrn': 'urn:li:school:3244736', 'entityUrn': 'urn:li:fs_miniSchool:3244736', 
'active': True, 'schoolName': 'Delhi Public School - India', 'trackingId': 'skI/DaQYTW23ZkqbA4FrPA==', 
'logoUrl': 'https://media-exp2.licdn.com/dms/image/C510BAQFJWBCrV-fjMg/company-logo_'}, 
'timePeriod': {'endDate': {'month': 3, 'year': 2024}, 'startDate': {'month': 3, 'year': 2009}}, 
'schoolName': 'Delhi Public School - India', 'schoolUrn': 'urn:li:fs_miniSchool:3244736'}]

as you can see, it returns all the relevant information related to the relevant field

retrieving company details

we can also retrieve information for any company. this is very similar to retrieving information from user profiles.

company=api.get_company('google')
print(company)

this will also give us a large dictionary. you can get information such as industries in which the company is involved in, staff count, headquarters, funding data, and many more parameters. they are as follows:

dict_keys(['staffingCompany', 'companyIndustries', 'callToAction', 
'staffCount', 'adsRule', 'companyEmployeesSearchPageUrl', 
'viewerFollowingJobsUpdates', 'staffCountRange', 'permissions', 'logo', 
'claimable', 'affiliatedCompaniesResolutionResults', 'specialities', 
'confirmedLocations', 'followingInfo', 'viewerEmployee', 
'affiliatedCompaniesWithEmployeesRollup', 'lcpTreatment', 'affiliatedCompaniesWithJobsRollup', 
'name', 'tagline', '$recipeType', 'fundingData', 'overviewPhoto', 'multiLocaleTaglines', 'description', 
'entityUrn', 'headquarter', 'showcasePagesResolutionResults', 'paidCompany', 'universalName',
 'viewerPendingAdministrator', 'companyPageUrl', 'viewerConnectedToAdministrator', 'affiliatedCompanies', 
 'dataVersion', 'companyType', 'coverPhoto', 'associatedHashtags', 'groups', 'url', 'showcasePages', 
 'claimableByViewer', 'jobSearchPageUrl', 'showcase', 'autoGenerated', 'backgroundCoverImage'])

remove connections

let's say you have a huge list of people you want to un-connect with. this can be done by the following snippet:

api.remove_connection('profile id')

if you have a huge database of profile urls, you can loop over them and then remove your connection with them using the above snippet. note that the profile id can be obtained by the value between the last 2 slashes in the profile url as mentioned earlier in this article

get conversation details

you can also retrieve ALL your conversations with everyone you have talked to on LinkedIn. this returns a huge dictionary with many elements. each of these elements contains data for each conversation in your inbox

print(api.get_conversations())

disadvantages and possible alternatives

however easy to use and cool this library sounds. it still has a couple of disadvantages which are listed below

sometimes LinkedIn might stop you with a captcha test. a solution to this issue is to log out and log back into your browser
this happens because of two-way factorization or rate limiting issues. The rough limit is said to be around 800 requests within a single session some possible alternatives could be trying to make your own scraper using selenium. this would require a lot of work and experimentation but it would also offer customizability and some extra features which would otherwise not be available in this library. these features can literally be anything since you are the creator of the scrapper and you can add/make anything of your choice! again it all comes down to what is the end result you want to achieve and use the tools available to you accordingly.

conclusion & project ideas

this library can be very handy and useful for a large scope of problems. it will save you loads of time and is very easy to use and you can set it up and start playing around with it in a matter of minutes! you can make cool projects like comparing famous people/profiles as a data science project since there is so much data available on LinkedIn. you can also try to make an ai model which can predict weather a person would get a job based on their profile. the model can be trained on employee data which can also be scrapped from LinkedIn itself. these were some fun project ideas. have fun coding :)

record VC audio in discord!

hardik singh — Sat, 16 Jul 2022 10:28:31 +0000

I have been working on this side project where I can record the audio of a vc with the help of discord.py

in this post ill talk about how did I achieve this. you can also check out the YouTube video i made about this.

before we get to the fun part, you will have to make and set up a bot via the discord developer portal. you will also need to have a fairly good amount of knowledge about discord.py and how it works!

now once you have a bot with a basic test command working, we can import the required modules!


 python
import discord
from discord.ext import commands
from discord.ext.audiorec import NativeVoiceClient #important!
from secrets import token #bot's secret token
import random

now to record audio in the vc, we need to first make the bot join the vc in which the user is, which can be done using this command:



@client.command()
async def join(ctx: commands.Context):
    channel: discord.VoiceChannel = ctx.author.voice.channel
    if ctx.voice_client is not None:
        return await ctx.voice_client.move_to(channel)
    await channel.connect(cls=NativeVoiceClient)
    await ctx.invoke(client.get_command('rec'))

as you can see, this command invokes a command named 'rec'. this 'rec' command is the main command which we will use to record the audio in the vc!



@client.command()
async def rec(ctx):
    ctx.voice_client.record(lambda e: print(f"Exception: {e}"))
    embedVar = discord.Embed(title="Started the Recording!",
                             description="use !stop to stop!", color=0x546e7a)
    await ctx.send(embed=embedVar)

we start the recording and then send an embed notifying the user that the recording has started!
the recording can be stopped and saved by using this command



@client.command()
async def stop(ctx: commands.Context):
    if not ctx.voice_client.is_recording():
        return
    await ctx.send(f'Stopping the Recording')

    wav_bytes = await ctx.voice_client.stop_record()

    name = str(random.randint(000000, 999999))
    with open(f'{name}.wav', 'wb') as f:
        f.write(wav_bytes)
    await ctx.voice_client.disconnect()

over here the bot basically gets all the audio data in the form of bytes and then writes al the data into a wav file(this file's name is randomly generated), which is then saved to the system
now since the bot's job is done, it leaves the vc!

some extras:
you can also add commands to make sure there is someone in the vc before the bot joins the vc, or uploading the audio file to a server on the cloud. all these points are discussed in my youtube video, feel free to check it out!

all the code for this project is available on github too!

realhardik18 / discord.py-voice-recorder

repo for recording the vc in an audio channel in discord.py

convert images to ASCII characters!

hardik singh — Sun, 15 May 2022 09:51:23 +0000

ever wanted to convert your image to Ascii characters? here is how you can achieve it!!

we will be converting this image of Mario, into ascii characters

step 1:- loading in the image and resizing

we resize the image since there are a LOT of pixels. so we resize it by 50%! this will make the final output smaller and we will not be compromising on the quality(since there are a lot of pixels to begin with!)

step 2:- converting the image to a pure black and white image

here we convert the image to a pure black and white image. we do this since text characters don't have color. now we can save this black and white data to a new image

here is what the image looks after the conversion!

step 3:- converting this black and white data to text characters

here we read in the data from the fresh black and white image which we saved in the last step. we also have a count variable which helps us to break the line for every time we finish a row section of the image
now we check the color of each of these pixels, if its black(0), we simply write in a "~". on the other hand if its white(255), we write in a "."
we now save this data into a text document!

here is how it looks like finally

I have resized my code editor so you all can see this masterpiece!

DEV Community: hardik singh

I created a tool that helps you learn from YouTube faster and more efficiently!

QuizMe.mp3: Turn Your Audio Lessons into Fun Quizzes

What I Built

Demo

Journey

realhardik18 / quizme.mp3

CuriShot: A Minimalist File Sharing and Organization Tool for Students

What I Built

Demo

My Code

realhardik18 / CuriShot

a webapp to organize and store code and screenshots take during my college lab classes

CuriShot: A Minimalist File Sharing and Organization Tool for Students

What I Built

Demo

My Code

More Details

More Details

How YOU can type faster

Understanding the keyboard

Now all you got to do is just practice!

Resources for typing faster

Introducing BookWheat, The ultimate twitter bookmark tool

Overview of My Submission

Submission Category:

Video Explainer of My Project!

Language Used

Link to Code

realhardik18 / Book-Wheat

a twitter bot which can categorically save your tweets and supports discord integration!

BookWheat, The ultimate twitter bookmark tool

Overview video!

How it works

How the data is stored:

the dog api!

overview

endpoints

learn to make a RESTful API with flask!

so what are we building here?

getting the data from the csv

now on to making the api!

final output

code

best resources for practicing data structures and algorithms

why even need dsa?

on to the resources!

conclusion

everything you need to know about carbon!

why do we need carbon in the first place?

plus points of carbon

goals of carbon-lang

project status

carbon-language / carbon-lang

Carbon Language's main repository: documents, design, implementation, and related tools. (NOTE: Carbon Language is experimental; see README)

Carbon Language: An experimental successor to C++

Why build Carbon?

LinkedIn API Python Library

LinkedIn API Python Library

why get data from LinkedIn with python in the first place?

introduction & advantages

retrieving profile information

retrieving company details

remove connections

get conversation details

disadvantages and possible alternatives

conclusion & project ideas

record VC audio in discord!

realhardik18 / discord.py-voice-recorder

repo for recording the vc in an audio channel in discord.py

convert images to ASCII characters!

ever wanted to convert your image to Ascii characters? here is how you can achieve it!!

step 1:- loading in the image and resizing

step 2:- converting the image to a pure black and white image

step 3:- converting this black and white data to text characters

here is how it looks like finally

Carbon Language:
An experimental successor to C++