- Duration: 40 min
- Platform: zoom
- Source: Read.ai 01JAR5A51VQFP87QDP5QPNPPZH (https://app.read.ai/analytics/meetings/01JAR5A51VQFP87QDP5QPNPPZH)
Transcript
[Rafael Bruno]: What’s up, guys? [UNKNOWN_SPEAKER]: Hey. [Mehul Jariwala]: Just waiting on Rafael. [Conference Room (Sid) - Speaker 1]: I think he and Sid were just talking, so give me a second. Hi, guys. Sorry for getting late. We were so lost in the conversation, we just forgot. [Michael Houck]: Yeah, I could tell you guys were in the details. Rafael is jumping in. [UNKNOWN_SPEAKER]: Great. [Rafael Bruno]: Hey, guys. Sweet. What’s up, man? I’m doing good. How about you? [Michael Houck]: Yeah doing solid Okay, I think let’s just jump right in I’m gonna share my screen Okay, so we’re not in a great spot, you know, you guys know I ate flat lines So not a great spot not a lot of progress the last few days even before the weekend obviously The good news is we’re not really finding many new bugs at this point, which is good, but by the end of this week it will have been one month since launch and we’ve literally done nothing this month except squash bugs, so not super great. But I’m optimistic that we can hopefully get through these this week so that it won’t go longer than a month. We really, really need to do that because it gets harder to justify spending the time on bugs the longer it takes. So yeah, let’s just make sure that we get through these this week. Some of these have been lingering in progress in the review for almost a week now. So I want to understand what the deal with those is first. I guess, Moho, you have a pretty light load right now. With you. What’s going on on your side? [Mehul Jariwala]: I have finished a part of the creator flow with both the tabs. I just moved to tickets in review. I just need your help. Probably I’ll try to send you this insert before I go into beta so that you can have the review it at once. And then after that, I was just fixing one bug that Rafael pointed out regarding the something I’ll yeah, that’s something I’ll work on it. And yeah, those once we that the one when we have that was done from the back end side, I need to make some few modifications on front end side in order to adapt the changes. Yeah. And after that, like group group feature that we are going to do or which what was the priority that we have after that? [Michael Houck]: Yeah. Okay. Side that makes sense after this bug I would say I would say the next thing is yeah moving on to well you can finish up the referrals thing by doing the mobile design and also the the banner behavior so right now we have one of the banners which is like make a referral right and I think it links like send an email or it links to like Tapfiliate or something like that. Right. We want that to link to the referrals page instead. So simple fix should take a couple of seconds. [Mehul Jariwala]: I’ll do that. And one more thing, create mobile design. I already finished that as well. And as a part of that, I thought we don’t have that ticket, so we can just cancel it for now. Yeah. [UNKNOWN_SPEAKER]: Okay. [Michael Houck]: So that was included in this. Great. Yeah. Yeah. Then on your side, just finish up this thing for the referrals. Um, and then we can move on to groups. I need to make some tweaks to the design for the groups page for inner circle specifically. So I will do that today so that you can get started on this by tomorrow. [Mehul Jariwala]: Oh, okay. Sure. And one more thing, right? So about the draw, are we going to have the, uh, draw our new design adopted for now, or we’ll pick up in of the later? Add a new account. You remember we design team has done the new UI where the side drawer is coming up from the right side. What is this for? The new drawer, the setting where when user add a new account, rather than showing a model, we were sliding the drawer from the left side. Maybe I’ll send you the link in so that you can just check on that. [Michael Houck]: Okay, yeah, send me the link and we can talk about that over Slack. Yeah, sure. Cool. Okay, I guess, Rafael, we’ll start with you on backend. [Rafael Bruno]: Where are we at with all these? Okay, so I finally think that I found the error that is causing all the wallet problems. Let me try to summarize is what I found with Sid. It seems that this was the process of creating tables. First, we get all the data from our table. Then we use that data to create the request table. Then we use the data from the ledger to create the charges table. After that, we created the wallet table and we populated that wallet. By the time that we populate the wallet and you are checking stuff, we figured out that there was an issue on the wallets and the numbers were wrong. So what he did, he used the data from the chargers and from the request table to fix the error on the wallet. But then a few days, I don’t know how many days later, We found out that the request and the charges table were not OK. So they should not serve as a source of truth. So basically, we used a bad data to fix a problem. And that generated another problem. So now we are working on a fix. And hopefully, I don’t know when Sid will go to sleep, but hopefully we can end this thing today. [Michael Houck]: So this is this is this one. This is not going to make this large because it’s been going on for a while. And how does this play into the other pending issues? So like engagements being marked incorrectly, the one that’s assigned to Rizwan, which is adjusting the scraper for the one hour time frame. We need to go back and do that for the past ones as well. [Rafael Bruno]: Like how do those play into this fix? Are we gonna need? About the running engagements to prep past broadcasts, I passed a API to Rizwan so he could be executing that on the weekend. We executed almost all of the pending broadcasts. We have just a few that I’m executing right now, so this is also being fixed. Hopefully, by the end of the day, I can eliminate that task, too. And the other task, where is the other engagements? No, this is the one that we are doing. [UNKNOWN_SPEAKER]: Yeah, just a scraper. [Rafael Bruno]: So this part, I think it’s from Reels 1 site. Because I don’t know. The next step here is on Rizwan. [Michael Houck]: We’re going to talk about this when we pass it to him. My question is more so when we do this, we will need to go back and look at the past engagements once again, because up till now, we’ve been passing all the engagements at once. They’ve been checking at the same one hour mark. So we’ll need to go back and make sure that that one hour mark is applicable to this specific request rather than the post as a whole, right? And so I’m wondering, what I’m wondering is, once we solve that, and once we rerun that for the past engagements, are we gonna have an issue here again? [Rafael Bruno]: And then- I don’t think that we can rerun those things there. Wait a minute. We can for Twitter, we can for LinkedIn. [Michael Houck]: For LinkedIn doesn’t show us- Yeah, no, no. [Rafael Bruno]: I was with a wrong thing on I thought that we are trying to get data from engaged super fast, but I was thinking wrong. Yes, things are ready here, so we can also do this. And there is nothing much to modify on the backend. Yeah, no, I know that. [Michael Houck]: But my question is, we are not done this ticket right now. There’s still work that Rizwan needs to do on 950. And we need to talk about what that work should be. This is not gonna be done today, probably, right? And so my question is, does 950 block us from doing 969? Because if we do 950 after, are we just gonna need to redo the issue with 969 afterwards? No, no, they are not related. Okay, good, great. That is a good thing. So, okay. [Rafael Bruno]: While I’m working on those issues, I’m trying to pick like small stuff so we can, like, and this backup, bugs backup, backup, not bugs, backlog that we have. Yeah, pretty much what I’ve been doing. Great. Okay, so we gotta do that faster, right? [Michael Houck]: Because we’re not seeing much go through. So this will be done today. This will be done today. And then we have to get through the rest of these this week. Basically. Okay. And Rafael, are you out on Friday? Or are you leaving like Friday night? [Rafael Bruno]: Friday by like, 6pm, I will probably won’t be online anymore. Okay, that’s fine. Let’s just get through these before then. [Michael Houck]: Things like sentry, maybe I can pass this to sit to set this up. But all the all the bugs definitely need as much of your time as possible on those as Sure. [UNKNOWN_SPEAKER]: Okay. [Michael Houck]: On that note, Sid, I’ll pass it to you. What’s on your plate? [UNKNOWN_SPEAKER]: Yeah, sure. [Conference Room (Sid) - Speaker 1]: So I was working, I mean, alongside the data re-migration, I was working on 838. That is completed. It’s that here on the screen. [UNKNOWN_SPEAKER]: Wait, where did 838 go? [UNKNOWN_SPEAKER]: No, sorry. [Conference Room (Sid) - Speaker 1]: I think it’s 930. My bad. 938. [Michael Houck]: I thought it was 38, too. But I guess not. Let me look through our chat, because I thought that was the number, too. [Michael Houck]: 898 and 935, was that it? [UNKNOWN_SPEAKER]: 935, yes, and 898 restart. [Conference Room (Sid) - Speaker 1]: Yeah, yeah, 898, my bad. 898, that’s right. So I mean, it’s first doing the data migration, but since I was also waiting for updates from Rafael, I thought I’d just move to other tickets. So 898 is completed. I just need to make an API call on the front end, but mostly it’s done. Data remigration, what we are doing, I and Rafael are doing. Once that is done, I’ll merge this on dev and we can test it there. If it looks good, we can also release it on production. Okay. [Michael Houck]: And then what’s the status of 940 and 829? Because these have been lingering for quite a while. [Conference Room (Sid) - Speaker 1]: Yeah. So 829, I think Rafael mentioned once that we need to do it with the help of Rizwan and Kafka. So Rafa, do you have any thoughts on that? When should we do it? [Rafael Bruno]: Because of the… We just need to create a payload and send that payload to the scraper. And the scraper will get that information and populate S3. [Conference Room (Sid) - Speaker 1]: So it will populate to S3? Yeah. [UNKNOWN_SPEAKER]: Okay. [Conference Room (Sid) - Speaker 1]: We just need to emit a payload to the topic. Yeah, fine. Fine. I’ll take care of this one since the migration thing is high priority. So once that is done, once Rafael confirms we are on good track with the data thing, I’ll take up this one, 929. I think that should be a quick one then. And 940, I think I fixed that one, but it seems that there’s still some issues. I’ll have to recheck 940 again. Because on production, I just checked it today. We were still having some issues with the followers. So I’ll try to get that data demigration out first, and then other tickets will be unblocked pretty much quickly. I think in a day or two, it should be unblocked from that point. Yeah. Okay. [Michael Houck]: We really need to move quick because we only have this week to get through all this stuff. And the workload here is, I think, considerably lighter than the full list for the sprint. So really need to get through the ones that are already assigned here so that we can add some more of the bugs and split them up more evenly. So if you have any questions or if you think it’s going to linger longer, please let me know. But I think this headshots one has been here for like two to three weeks now. And this one has been here for two to three weeks also. So we really, really need to solve those. If there’s an issue, we’ll ask the longer than tomorrow, please let me know. [UNKNOWN_SPEAKER]: Sure. [Conference Room (Sid) - Speaker 1]: I mean, once this is done, I’ll let you know once the data re-migration thing is completed. [Michael Houck]: Right, right. [Conference Room (Sid) - Speaker 1]: But my point is that if that’s going to be completed today, it should be completed tomorrow. Correct, correct. [Michael Houck]: I’ll let you know if it is done today itself. [UNKNOWN_SPEAKER]: Okay, great. [Mehul Jariwala]: Okay, and then Rizwan, let’s go over to you. [Conference Room (Rizwan Hameed) - Speaker 2]: Yeah, so I have added my input over this one, which is to access the Scripter. And after that, I’ve started working on the engagement URL thing. So we need to finalize what will be done on this part for 950. And other than that, I’ve been working on the engagement UI thing. So these are the two updates on my side. Other than that, we have made a lot of changes in the last week as well, and executed a lot of engagements as well. [Conference Room (Rizwan Hameed) - Speaker 1]: So that was on the last week. So this is the update so far. [Michael Houck]: Is there anything you want to talk through here? You left some comments earlier, and I replied to you on here. [Conference Room (Rizwan Hameed) - Speaker 1]: Yeah, I have gone through to the comments as well. And now I need to like, you know, discuss what’s the final approach. I don’t need to do anything on that because I’m not, I didn’t have any better idea other than sending the posts, same post again with the different appifiers because we need to limit the use of the new accounts as well. So if there is any suggestions how we can deal with it, that would be much appreciated. [UNKNOWN_SPEAKER]: Otherwise, uh, don’t have a way forward. [Conference Room (Rizwan Hameed) - Speaker 2]: Well, okay. Does it count for multiple views or requests or whatever? [Michael Houck]: When we view the same post, like different times. So if I click on a post and I view it to get one person’s engagement, but then I click on those later to get, um, a different person’s engagement. Is that, does that count too against our, our limit or does that only count one? [Conference Room (Rizwan Hameed) - Speaker 1]: It’s two. [UNKNOWN_SPEAKER]: It’s two? Yes. Ah, that sucks. [Conference Room (Rizwan Hameed) - Speaker 1]: Yeah. That’s why I recommended, like, you know, the best solution that we can do is just increase the limit. [Rafael Bruno]: I don’t know how is the… Sorry, Rizwan, to interrupt. You can continue. No, no, I’m done. [Conference Room (Rizwan Hameed) - Speaker 1]: You can continue. Please go ahead. [Rafael Bruno]: Okay, I don’t know how is the life cycle of the bot, but is there a way to, like, make the bot alive so that instead of losing some sort of state, that bot could go back to another post without counting? I’m not sure if that’s possible. [Conference Room (Rizwan Hameed) - Speaker 1]: It is, but it’s going to raise other issues as well. When we keep the bot Chrome session open for this amount of time, it gets scratched as well due to the memory and the caches issues as well. So those issues becomes the hard blocker as well in this scenario. Plus, again, even if we keep the one session of the browser, we are still counting these, we are still visiting different profiles, right? Different posts, pay links, which are counted as a, which are being tracked by the LinkedIn as well. Doesn’t matter so right now what I do is like you know I keep the similar information in every event in every post right for the browser so browser knows like you know it’s a consistent session every time so it’s not a issue on that part the issue is on the LinkedIn side like you know if we are visiting a post multiple times or like a let’s go with the example of the profile if we are visiting X number of profiles in a day so LinkedIn is going to block our profile or is the warning doesn’t matter how many times the browser is open it’s going to count the event against the logged in profile how many times this user visits the number of profiles so such types of issues or the limitations is on the linkedin side it’s not on the browser side well okay just to confirm they don’t need to visit the profiles in this case they [Michael Houck]: just need to visit the posts and it’s the same link clicking each time it’s not like it’s a different URL. [UNKNOWN_SPEAKER]: Yeah. [Michael Houck]: Is that still an issue? Still the same issue? [Conference Room (Rizwan Hameed) - Speaker 1]: Yeah, because I’m receiving the post engagement either like it’s the same post, but with different number of amplifiers. So I’m receiving this post at different times or different as a different events, right? So one event is for one post with the given Yeah. [Michael Houck]: So the, the, an example would be, uh, someone shares a broadcast with the network. The algorithm says, okay, we’re going to share it with one amplifier right now, but this other amplifier match for it. Isn’t coming online for 20 minutes. And so we’re going to share it with them in 20 minutes. Right. So what that means is we have two different one hour timers going. One starts at the beginning of the shared and the second one starts 20 minutes later. So after that, we’re going to share it. One hour you need to check the first creator’s engagement and after one hour and 20 minutes we need to check the second the creator’s engagement right so we go to that same post twice but it’s the same url we’re [Conference Room (Rizwan Hameed) - Speaker 1]: not clicking on any profiles or anything like that just going to the same post twice is that that is that still an issue yeah just because uh it will keep the bot uh like busy with the same post we are running a on the board. So it is going to keep the board busy and keep the board waiting for the same post. If I understand it right. So like, you know, as you are saying, we need to keep, we need to stay on the same post and keep waiting until the new time matches. And then we run the engagement again, engagement matching again. [UNKNOWN_SPEAKER]: That’s right. [Michael Houck]: Either staying on the post or leaving and coming back either one. [UNKNOWN_SPEAKER]: Yeah. [Conference Room (Rizwan Hameed) - Speaker 1]: Then leaving and coming back is what is happening right now. But it’s not real time. It’s going to take some time because we are running each post one by one. So it gets skewed and will be processed later. No, no, but maybe we are missing something here. [Michael Houck]: Yeah, I think we’re talking about different things here. I’m sorry, there’s a helicopter flying by my house right now. So OK, very simply, one post is shared. It’s matched with two amplifiers. One amplifier is available now, so we send it to them. One amplifier signs on in 20 minutes. That’s when their availability hour starts. So we send it to them in 20 minutes. Then one hour from the time that the post was shared with the network, we have to go check on that first creator’s engagement, right? And then 20 minutes after that, we need to check on the second creator’s engagement. So we can either have the bot go to at one hour and hang around for 20 minutes, and then check it again after that time, or we can have the bot go in an hour and then leave and then click the same post URL to go back to the same post 20 minutes after that to check for the second. So what I need to know is, does that count as two visits and two against our 30, or does that just count for one since it’s the same URL? [Conference Room (Rizwan Hameed) - Speaker 1]: It counts as two because every time we refresh the page, we refresh the URL or we reload it, right? It’s going to count as two visits of the profile or the post. Okay. [Michael Houck]: So even though we’re not visiting a profile, it still counts. Yeah. [Conference Room (Rizwan Hameed) - Speaker 1]: And I’m not sure right now, I’m not sure about the number of limit, the counts limitation that LinkedIn has for the post. We do have some random about the 30 or 40 for the profiles, like a 50. Plus something, but not sure about the count of the post visits. [UNKNOWN_SPEAKER]: Oh, I’m not sure if, sorry. Go ahead, Rafa. You can go on. [Michael Houck]: So, okay, so we know that it’s 30 profiles. We don’t know if there’s a higher limit for the number of posts we can visit, right? Yeah. [Conference Room (Rizwan Hameed) - Speaker 1]: We have, based on some random RLDs, like, you know, we’re going through the different blogs, the limitations or the different articles, so we decided to go with around 30 posts to visit, like, you know, not to exceed the, not to bust it on the living side. And it’s going good so far right now. [Michael Houck]: Sorry, you’re, you’re breaking up the audio quality. I can’t, I can’t. Is it better now? [UNKNOWN_SPEAKER]: Yeah. Yeah. [Conference Room (Rizwan Hameed) - Speaker 1]: So based on the different articles and the R&Ds, so we decided to go with the existing count, like visiting around 30 posts per day. And it’s going good so far. It’s been a month and it’s doing properly fine. And there isn’t any concern over the account security. So. [UNKNOWN_SPEAKER]: How many are we doing per day right now? [Michael Houck]: Because maybe like within 30 or would we need to increase it or buy more accounts or like how many are we doing right now? [Conference Room (Rizwan Hameed) - Speaker 1]: Yeah, right now per account it’s around 30. Like it can be I have a random limit like you know, day, like it’s between 28 to 33. So it’s a random count per day. [UNKNOWN_SPEAKER]: Okay. [Michael Houck]: So we would need to either add more accounts or increase the limit if we wanted to implement this, right? [Conference Room (Rizwan Hameed) - Speaker 1]: Yeah, definitely. Because as I said earlier, the bot performs in a sequential execution, one by one. And now we have implemented for the entire to distribute the post on three different machines. So that’s doing so far, but these has three accounts and it can process around 100 posts on daily basis. So to increase the limit for this approach, like you know, there can be multiple events for the same post with different amounts of amplifiers. We need more accounts and machines as well. [Michael Houck]: So right now we can process 100 of these events per day is what you’re saying, but we need to increase that. [Conference Room (Rizwan Hameed) - Speaker 1]: Yes, it is what is happening right now. [Michael Houck]: Any thoughts, Rafael? [UNKNOWN_SPEAKER]: I thought about something. [Rafael Bruno]: Correct me if I’m wrong, but I don’t know if it’s the same idea that I gave before. But thinking about a browser, when I open a browser, I could have several tabs. Is it possible for you to create some sort of tab and you keep, I think it’s the same idea of keep waiting on the post, but having something stored on a tab so you can go back later. [Conference Room (Rizwan Hameed) - Speaker 1]: Yeah, so yes, that is doable, but it’s gonna, again, like, you know, however, let’s suppose if there are 10 plus or 20 plus posts for which I have to So it’s going to overflow the memory of the browser as well, right? If there are multiple tabs which are being opened and in the queue right now, plus this needs a lot of core changes in the board execution as well, because board needs to keep checking which tab to check after how many times, after how many amount of time. So this is going to be a huge implementation in the core of the board as well. [Rafael Bruno]: So if I understood correctly, we have at one side, we have to pay for more accounts so we can continue to do that. Or on the other side, we have to pay on our infrastructure, on AWS, to have better machines to support that new idea, right? Yeah. Are you able to quantify how much How much would it cost each option? [Conference Room (Rizwan Hameed) - Speaker 1]: I can’t share it right now, but I can look into it and can get back to you after that. [Michael Houck]: Yeah, we know how much it would cost for the accounts. I think it’s like, I think it’s like 125 per month per account or something like that. It’s pretty expensive. [UNKNOWN_SPEAKER]: Yeah. [Michael Houck]: So if the infrastructure costs me more, obviously what we’ll do, we have to do, but we definitely need to make this change one way or the other because every single broadcast is basically going to be wrong right now, because we’re not tracking. Creators at different times, if they get matched at different times, so. Yeah, this is like pretty urgent, so I guess everyone, let us know what you find on this today, Rafael, if you have more thoughts, chat with everyone about it. [Rafael Bruno]: Could you please explain again the problem of shadowban of It’s like every action that we do could cause a shadowbomb if we use the API. [Michael Houck]: Yeah, sorry, I do have to run to another meeting in a second. Go ahead. But just really quickly, basically, if we use the API, then LinkedIn or X won’t be super happy that some of our users are paying for engagement on their posts. And so they might reduce the visibility of that user’s posts if they get flagged for it. Now, getting flagged is pretty low likelihood, but if they do get flagged, they could reduce the visibility of their posts. And then that would be really bad for our business because users would be saying, I use this product and it fucked my shit up, you know? So, um, we can’t have that happen, which is why we don’t use the API. Okay. Got it. Okay. I got a hop, but, uh, you guys should keep talking about the scraper thing because, uh, we need to solve that. [Rafael Bruno]: Um, ASAP. Okay. [UNKNOWN_SPEAKER]: Okay. [Michael Houck]: I will. I’ll hop out. [Rafael Bruno]: I’ll see you guys later. See you. So guys, do any of you have any other suggestions of what we could do here? Basically, let me try to summarize the problem. As we scale the business, we will have more people more accounts on our platform. And it’s really bad if we have to pay for more LinkedIn accounts to keep checking on the requests. And as we scale, the number of engagement confirmation that we need to do will increase. So it will be good if we have a option that we can optimize the way that we scrap the information from the posts. And as Rizwan said, we have a limit of how many, if I’m not mistaken, how many profiles one account can visit per day. It’s about 30 profiles. So if any of you have any idea of how we can implement something better or how we can change the flow to other this, we are open to suggestions. [Conference Room (Sid) - Speaker 2]: From what I know Drisvan, if you upgrade to LinkedIn premium, the number of accounts you can visit will either increase by three to four times easily, or I think it is unlimited. I don’t know. But I used to get this problem long time ago. And then for some other reasons, I upgraded to premium. And after that, I never came across that problem of, of the number of accounts I can see in a month. [UNKNOWN_SPEAKER]: Maybe you can give that a try. [Conference Room (Sid) - Speaker 2]: Because LinkedIn premium, I’m sure at least you’ll get rid of the problem. Now that there could be that the number of profiles you can visit will increase. For example, if it is 30 now, after upgrading, it may become 100 or it may become unlimited. I don’t know that one. You can double check on the pricing or something. But at least that will eliminate the problem for now for today. By the way, how many accounts do we have? [Conference Room (Rizwan Hameed) - Speaker 2]: I believe we have our own six or seven accounts right now for the link. [Conference Room (Sid) - Speaker 2]: And are we using any service like third party service for those accounts? [Conference Room (Rizwan Hameed) - Speaker 2]: Yes, we are. So which one is the link? Yeah, so it’s called accountify. LinkedIn has added a lot of different validations for the new signups and also detects or tracks the activity of the new accounts as well. So those accounts were being detected and blocked. That’s why we went for a third-party service, which has pure and mature accounts and their accounts, identified accounts as well. [Conference Room (Sid) - Speaker 1]: I think that might do it. [Conference Room (Rizwan Hameed) - Speaker 2]: The premium thing that should do it. [Conference Room (Sid) - Speaker 1]: In my opinion, you can give that a try. Yeah, it’s a good idea. We can give a try. [Rafael Bruno]: You can suggest that to Hulk and he can decide to upgrade one of one account and see the results. [UNKNOWN_SPEAKER]: Yeah, but we go. [Conference Room (Rizwan Hameed) - Speaker 2]: Yeah, that’s a good suggestion. And definitely we should bring this to the Hawks table as well. But the main questions come back to the board as well about the sticker. Like, you know, we have different times for different intervals for different amplifiers for the same post. [Rafael Bruno]: So. Yeah, so that if we have, let’s say that’s its idea that it will pay and we will have one account with unlimited access. It can visit how many posts it wants by day. [UNKNOWN_SPEAKER]: We would solve that problem. [Conference Room (Rizwan Hameed) - Speaker 2]: Right. So you mean you will be sending a different event for different intervals, right, for the same post? [Rafael Bruno]: Yeah, it’s the It’s the events that we already send like super fast, in time, and late. [UNKNOWN_SPEAKER]: Right. [Conference Room (Rizwan Hameed) - Speaker 2]: Yeah, I think that’s a very good suggestion. And any other suggestions, guys? Maybe that can ever add up a huge list with a lot of better suggestions. [UNKNOWN_SPEAKER]: Anyone? Mehul? [Rafael Bruno]: Ben? [Mehul Jariwala]: Yeah, we’re just thinking through it. [Rafael Bruno]: I’m trying to look at the LinkedIn premium because it is a good idea in theory. I just wonder if it would still work, like if it still solves the core problem. In theory, it does. Okay, so besides that, is there any other topic that we need to discuss today? Okay, so let’s give that idea to Raul. And uh he can decide on that okay yeah sure I’ll add the details over the ticket and we’ll also spend some time over the r&d as well okay to get more details about it okay so uh good talk see you guys bye see you have a good