Reddit accounts farm

newmersedez

Newbie
Joined
Nov 13, 2023
Messages
24
Reaction score
11
Hello BHW users! This is my first post on this platform, so I would like to share with you my experience in Reddit account farming.

My journey started about 6 month ago, when my friend told me about great opportunity to earn some money with programming. It seemed very interesting to me as a Computer Science university graduate without any income stream or 9-5 job. I'd been actively programming in Python and C# for 2 years, so this deal made me very interested.

The main goal was to develop a private Telegram bot with comfortable design for me and my friend to create and boost Reddit accounts. Algorithm is pretty simple:
  1. You manually buy a residential proxy for each account,
  2. Bot creates account with given username, password and proxy and warms it up for 3 days
  3. Bot writes comments with ChatGPT
  4. Bot submits manually scheduled posts.
I'd chosen Python as a programming language because it has such a beautiful (now i can't say that) module as Selenium. I'd tried it and of course it failed - accounts was getting banned right after creating. I continued to research and found Reddit API and module PRAW, but after tests result was the same. I gave up on the idea for a month.

After a little break I continued development with new programming language for me - TypeScript. I found out that Typescript and Puppeeter-stealth plugin might be a good solution of account bans problem. I'd developer an absolutely new version of my bot that was copying real user behaviour:
  1. I use residential proxy for each account
  2. Bot creates account with given username, password, warm up it for 3 days
  3. Bot emulate Android or iPhone device
  4. I use cookies for authentication so I don't need to authenticate again with every action, look very suspicious I guess
  5. In random time bot goes to main page or subreddit, scroll, "read" posts, upvote some posts, leave comments
  6. In random time bot goes to subreddits and writes comments using ChatGPT
  7. I manually schedule posts, bot submits them in given time
I had high hopes for this method, tried to buy aged accounts and different proxy providers, the result is the same - accounts are getting banned after a day of boosting (after 3 days of warm up). Now I have like 100% developed Telegram bot for Reddit accounts farming, but the main goal is not achieved, because accounts are getting banned, so bot is useless. Currently I stopped work on this project because of prolonged failures and lack of motivation, but I hope my story is being read by people who have conquered this problem and are willing to give helpful advice, tell me what I'm doing wrong.

Thanks for your attention, would welcome questions!
 
Hey bro I'm actually doing the same thing, when I read you I can see all my project hehe nice to see I was not alone , bro if you try to mimic as much as possible human behaviour and don't be greedy it will work! , how are you managing your fingerprints? Most of the time you get banned because: bad proxies / bad fingerprints / being greedy on action.
 
I'm in the process of doing something similar and I figured using AdsPower, Selenium, high quality residential proxies and aged accounts would work but the accounts are getting banned after a day or two. There's certainly ways to get this done but I think it might have to do with CQS. If anyone wants to discuss this feel free to pm me.
 
Now Reddit is banning accounts left to right. I think now they are using AI to detect fake accounts.
You will have to do a lot of testing.
 
Tips to prevent banned on reddit accounts
1. Use updated automated browsers with fingerprints and profile the accounts for later use
2. Once account is created let it rest for 3 days before commenting
3. Carry out less activites on the fresh accounts after 3 days eg 1 comment per day for atleast a week then increase comment count gradually to build trust on reddit
4. You need good GPT3 prompt to reply comment on less strict subreddit else some users will notice you use AI and report your account.
5. Good proxies must be used 4g/5g is highly recommended
6 Use 100% human emulation

I was also getting banned till i figure out the issue and had thousdands of acc with zero ban rate
 
Let's say someone had access to a "reddit farm".. what would be the best way to monetize it?
 
can you share doing "Bot emulate Android or iPhone device" code with selenium?
thank you
Actually I don't use Selenium anymore, app developed with Puppeeter, it allows you emulate any known device only with single line of code

Hey bro I'm actually doing the same thing, when I read you I can see all my project hehe nice to see I was not alone , bro if you try to mimic as much as possible human behaviour and don't be greedy it will work! , how are you managing your fingerprints? Most of the time you get banned because: bad proxies / bad fingerprints / being greedy on action.
I use puppeeter-stealth, guess it works with fingerprints instead of me + I emulate android/iPhone.
Maybe I'm doing it wrong, seems like it is

Bro what are you doing with the bots? you have to go very slowly
After 3-days warmup bot starts writing comments to AskReddit. Just about 10-14 comments per day with 1-2 hours interval between them. As soon as accounts reaches 30+ comment karma, bot is ready for posting.
After your comment it seems like I should comment less often to avoid suspicion.

Tips to prevent banned on reddit accounts
1. Use updated automated browsers with fingerprints and profile the accounts for later use
2. Once account is created let it rest for 3 days before commenting
3. Carry out less activites on the fresh accounts after 3 days eg 1 comment per day for atleast a week then increase comment count gradually to build trust on reddit
4. You need good GPT3 prompt to reply comment on less strict subreddit else some users will notice you use AI and report your account.
5. Good proxies must be used 4g/5g is highly recommended
6 Use 100% human emulation

I was also getting banned till i figure out the issue and had thousdands of acc with zero ban rate
I really appreciate your help! I think I successfully follow the rules 1-5, but 6 rule is too broad. What would you recommend to advance bot behaviour for real human emulation?
 
3 days old accounts won't survive or reddit with a bot
I've tried 6+ month aged accounts, nothing happened. I bet the reason in bot behavour and account trust, not in age, but thank you for your comment, it forced me on some ideas.

Hey bro I'm actually doing the same thing, when I read you I can see all my project hehe nice to see I was not alone , bro if you try to mimic as much as possible human behaviour and don't be greedy it will work! , how are you managing your fingerprints? Most of the time you get banned because: bad proxies / bad fingerprints / being greedy on action.
What tech stack would you recommend for such software development? Currently I'm using only pure Puppeeter with stealth plugins and Typescript. Can I for example use code to automate AdsPower for better fingerprints management?

Tips to prevent banned on reddit accounts
1. Use updated automated browsers with fingerprints and profile the accounts for later use
2. Once account is created let it rest for 3 days before commenting
3. Carry out less activites on the fresh accounts after 3 days eg 1 comment per day for atleast a week then increase comment count gradually to build trust on reddit
4. You need good GPT3 prompt to reply comment on less strict subreddit else some users will notice you use AI and report your account.
5. Good proxies must be used 4g/5g is highly recommended
6 Use 100% human emulation

I was also getting banned till i figure out the issue and had thousdands of acc with zero ban rate
I use AdsPower for creating accounts, also use unique proxy for each account.
Which tools would you recommend for developing? I've already used Reddit API (PRAW), Selenium with undetected chromedriver, Puppeeter with stealth plugins.. Maybe there are some tools which I actually don't know about
 
I use AdsPower for creating accounts, also use unique proxy for each account.
Which tools would you recommend for developing? I've already used Reddit API (PRAW), Selenium with undetected chromedriver, Puppeeter with stealth plugins.. Maybe there are some tools which I actually don't know about
What proxies do you use? thanks
 
Undetected chromedriver or puppeteer with stealth modules aren't going to help you. You need to develop private anti-detect solution that isn't publicly available on github.
 
Undetected chromedriver or puppeteer with stealth modules aren't going to help you. You need to develop private anti-detect solution that isn't publicly available on github.
Oh, it sounds really difficult. All this time I've been thinking that the main reason of my failure it that my action on account solution are suspicious. So, all this time I've been solving wrong problem. It there any easier way to solve it?
 
Tips to prevent banned on reddit accounts
1. Use updated automated browsers with fingerprints and profile the accounts for later use
2. Once account is created let it rest for 3 days before commenting
3. Carry out less activites on the fresh accounts after 3 days eg 1 comment per day for atleast a week then increase comment count gradually to build trust on reddit
4. You need good GPT3 prompt to reply comment on less strict subreddit else some users will notice you use AI and report your account.
5. Good proxies must be used 4g/5g is highly recommended
6 Use 100% human emulation

I was also getting banned till i figure out the issue and had thousdands of acc with zero ban rate
Bro , what programming language and library did you used for your bot?
 
What tech stack would you recommend for such software development? Currently I'm using only pure Puppeeter with stealth plugins and Typescript. Can I for example use code to automate AdsPower for better fingerprints management?
I use playwright and Gologin as browser also I think that commenting +10 post with new account is way to more, its better to post only 2-3 on multiples accounts
 
Back
Top