Singapore mobile proxy farm journey

Hi everyone, I recently started a mobile farm and would like to share my experience as well as interact with the BHW community.

I purchased a SIMBA card in singapore, which is our 4th biggest (and also smallest) telco and it seems like i im unable to rotate the ip when i click on the rotation link in proxy smart. it works for the other 3 telcos (singtel, starhub, m1) though. Does this mean that SIMBA sim cards cannot be used as a mobile proxy? or is it still useful but maybe certain functions are limited?

I also noticed that some of my m1 proxies seem to be going down more frequently than the rest. How serious is this problem? i heard that going down is normal for mobile proxies and they need to be restarted regularly. so how bad is it that it goes down more than normal? is the proxy still saleable and useful?

I'm happy to attach pictures but not sure how to do so. my setup is a optilex5050 with 16gb ram and sipolar 32 port usb hubs and huawei 3372h dongles

I'm also looking for information on business plans from singtel, starhub or m1 or their MVNOs. SIMBA is offering good business plans but the IP is not rotatable. As for the big 3, the advertised plans seems to be too expensive (5-10x normal user rate). but i noticed some MVNOs have this contact us for a chat function. do you have experience buying a data line as a business in singapore? i saw circles.life has one but im not sure whether they are going to offer it at a good rate.

---
I also noticed that there are real mobile phones available for rental at $250/month. How is that market different from mobile proxies? i have some spare phones as well, will they work or are they too slow? is it worth it to purchase a phone just for rental? its us200 here for a pixel 2, but only us20 for a dongle. the software cost per dongle is US1/month while the phone is US10/month. 10x difference here

so this new m300z LTE Modem is here, it looks quite bulky like 3 usual modems stuck together. they provided this usb c to c cable.... but my usb port is the regular A and usb2.0. i tried some of my A to C wires but they didnt connect to the phone. not sure how to do it.

in the mean time i purchased 5-6 models from separate sellers in aliexpress and most of them worked! now i can get these modems at less than US32 from at least 3 different suppliers.

i have more details about the plans now. at starhub they offer SGD22 for 150GB. comparing to a personal plan from other networks, i can get as high as 300GB for SGD10, so this business pricing is about 4x worse than personal market rate. if this business is to continue, it will have to be based on the business rate. maybe something like US60 per 150GB per month. they do not offer unlimited data. on my personal phone that i use regularly as a hotspot and watch movies, i cant finish even 100GB. but i guess on the modem it will be used way more but i think its still hard to hit 150gb unless you are downloading videos.

one of my buyers asked me to try out the m300z again... so i used one of the USB C to A converter plugs and it turns out that it is able to connect with the server. it could even tell exactly what provider it is, not just the parent network. this is going at 60% the price of a e3372h and looks promising. others say it can be very fast, but it looks like in singapore, the speed is limited anyway so its potential is not fully realized.

i might buy 20 of these USB-C to A because im expecting alot of new modems that i test will come with a USB-C plug. this setup is only for 4G i think? but 4G is here to stay at least for another 5 years.

my key buyer just asked to increase the number of modems by 10, which effectively uses up all the test sim cards and modems i bought. so i have to buy another 10-20 more for testing.

im also testing out using an actual phone to make a proxy. iproxy is charging USD10 per line which costs as much as the mobile subscription. i wonder if proxysmart can be used or any other cheaper software? this is a 5G setup and i think it should at least sell for US100/month/line.

i was about to buy over the equipment from another farm that had alot of e3372h-153. i happen to have one of them and it turns out that the signal bands are lesser on 153. the signal is like half, so unfortunately it would not make the cut for long-term-use server equipment

quite alot of my orders for the e3372h has been cancelled as they do not have enough stock. has this been discontinued already?

so this pushes me more to explore other options like the m300z. it seems like on aliexpress the e3372h and m300z are from the same group of sellers.

the m300z comes with USB C and i need to buy a adapter to turn it into USB A in order to plug into my sipolar hubs. i bought 3 of these adapters for testing.

i also looked around and found sipolar also has USB C hubs... but only 2 versions, so i bought one for testing.

there is another buyer who wants 3 lines, one of each carrier in singapore, m1, starhub and singtel.

yesterday there was a surge in use in server resources, basically maxxing out the memory, CPU, disk io etc. so i had to reset it. but what caused the surge? no one seems to know. i will have to be on standby to fix this when it happens

hello everyone, everything was quite dry in the first 2 weeks, and then suddenly a flurry of leads came in and bought out my inventory. looks like they are all from blackhatworld forum. I think the age of the buy sell thread and replies may have increased consumer confidence in my services. maybe I should get a sticky thread to further advertise? I dont want things to move too quickly. the modem purchases takes 2-3weeks to arrive, which is my decision time frame. I want to see maybe 10 rebills from separate customers before I plonk more money into this - at the same time maybe I should purchase more advertising. my current setup could fit up to 3 servers which is 132 lines which will be the threshold for me to expand the farm to another physical location.


I have bought out the cheaper m300z modem, and whats remaining now is the e3372-607s. I have a 153 and it has limited bands.
the remaining e3372 are priced higher but it looks like im going to have to accept that price. I believe in investing in the best (or 2nd best) equipment, tools and software for my life and work.

I have packed the server setup into a neat package that looks like it fits in a small box so as not to attract wanted attention. the m300z are a handful to manage and maybe I will have to purchase a organiser to keep them neatly. I would much prefer the e3372. small and easy to move around and doesn't drop all over the place.

after looking around it seems like I will have to accept the higher rates for business mobile data plans. there are no alternative mobile plans for businesses thats close to individual mobile data plans.

I have started to receive re-bills which greatly increases my confidence level, making me think this is something long term worth investing in.

I have also applied for a static ip for my home so that I can save on VPS costs monthly. MyRepublic only charges a one time fee instead of monthly fees like VPS. im glad I got the gamer's plan which is basically the best u can get in singapore.

as for the software, looks like it gets cheaper the more licences I purchase. so at this time I start to enjoy some economies of scale. I am able to fully utilise the capacity of the server, hardware etc.

the tech here feels shaky. im not a coder and it seems like there are alot of technical setups that I have to depend on others to provide information. what if I dont get the right person? what if there is a problem I cannot solve? how can I learn more and faster? is there like a mastermind group for this or something?

Great update, keep us posting!

I just received feedback that my m1 SIM cards often go down and some of my clients dont want it anymore. but at the same time, someone needs the m1 line to be specifically from this operator. also, compared to starhub and singtel, the data cost is like 50%. I will probably still offer m1 maybe at a slightly reduced price compared to starhub.

the plans I am signing up for now are actually 5G plans. but my current setup is a 4G dongle. 5G setup seems to be really expensive. and is 5G required for most operations? my understanding is that most of the data used is of low speeds. how long can the 4G dongle hold up? in singapore the 3G network has just been decommissioned. this means 4G is the default and it should remain that way for at least 10 years.

e3372 dongles are running out online, so I'm testing out other variations like the MF883V. the previous few I tested did not work out even though my software supplier mentioned it was compatible on his website.

I have also have opened up a company for this and setting up the finances and logistics. there is a 9% GST applied to all products but luckily that only applies to companies with more than SGD$1,000,000 in annual revenue. this 9% is really a big hit. they used to not enforce this for digital or e-commerce products, but recently stepped up their enforcement measures. I intend to keep each company revenue under 1M in order to reduce hassle and compliance. I think its also better to have different companies for different location so that in case one gets any compliance issues, the rest can still continue to run.

starhub has agreed to give me a no contract sim only plan, which is really a huge plus. singtel on the other hand is not very responsive and I purchased a 12-month contract just to get things started and hopefully get their attention. I think singtel should also be able to offer a lower term contract. this is important because if the company closes and is unable to fulfil this financial obligation, Singtel will send legal letters targeted at the individual to recover the money. so really this is a personal line with additional company cover.

I think I will slowly grow this to 132 lines max at my current rented apartment before investing any more cash for another location. I intend to use my parent's place to host this setup and I think I can host up to a maximum of about 600 lines. I have also lined up a location in Malaysia, but the profit margins there are smaller so I will still prioritise the singaporean farm. this is also when I will invest a larger amount into the equipment and software and advertising.


I have been thinking of my pricing strategy - and I think it's easier to follow an existing prescribed format, which I call the singapore way. most things in singapore are world class and really good. its not always the best. but it gives the most value per dollar. in terms of price, prices of essentials are generally affordable, but the prices of luxury are generally higher than other countries.
my strategy will be
1. to provide high quality, high uptime and stability (aiming to be 3rd in the world, not the first)
2. to offer it at an affordable price that beats most other proxies offered in a similar category.

and I think how it works from here is to let word of mouth to do the marketing instead of active advertising.
I intend to make my profit from efficiency in operations, the way in which setup and pack the dongles, the use of electricity, cost of rental of physical space etc.

and hopefully I get a loyal group of recurring clients that are happy with the quality and the price.

just remember there is no selling / promoting
in general chat threads.

providing telecommunication services in singapore requires a licence from IMDA (Infocomm Media Development Authority). this is true even though I do not intend to have any customers in singapore. I am currently going through the the process of finding out exactly which licence I need. this is done at the advice of my wife, but I think I would have reached this status myself anyway. I believe that I can run businesses lawfully as long as there is a valid need or concern.

I just signed up for a week's sticky ad and I'm excited to see how it works. when I had a free account, I already had people reaching out to me. and then when I got a Jr VIP account, even more. the sticky ad seems to be the next step after the Jr VIP I think. and then after that maybe the text ads and the banners. I expect the ROI to go down though, as it's less and less intentional. those who reached out right at the start are really interested and reading the chat. and then those who want to buy will go to the proxies for sale page, where a buy sell thread would work very well. for as long as the ROI is positive I think I can keep investing as long as there is a healthy margin.

I decided to set up another 2 more farms. 1 at my parent's place and 1 in Malaysia. ideally this can go on concurrently. my initial research in Malaysia says that I will need to incorporate a company in Malaysia. the profit margin in Malaysia is lower but the cost of equipment is the same. I think there is a cap to the business in singapore. I think there is a critical volume where the proxies can be differentiated from other users once it's too much. after all this is about processing data. maybe there will be some changes in IPv6 that makes the IPv4 I'm using less effective. this will happen as we progress from 4G to 5G. but I think this is a matter that will take at least 5 years to change and maybe another 10 years for a full transition to 5G, making my setup worthless. at some point in time, everyone will be on ipv6 and there will be no more IPv4. but I think it will take 15 years. look at the USB-A plug. its still around.

I am currently writing up a business plan for my farms. I have written maybe more than 50 business plans and I feel that it helps to give clarity and direction. the plan is alive, which means its constantly being updated on a weekly basis as things change and as I learn more things. I also keep a log of what I do so that I can look back and reflect to study the effects of my actions.

my internet also went down recently for 4-5 days already, and im concerned that the farm setup might have something to do with it. the modem does not work on the 10G cable, but somehow my farm is still working without issues (when modem is connected to router and router connected to server by lan cable)


I have also decided that my operations will be very lean. manpower will be just me if possible. this business doesnt really seem to require manpower other than setup and customer service, which usually is quite short. I dont think I enjoy managing other people and it looks like I dont have to anyway. whoever wants to manage this will also need a certain level of technical knowledge which does not make sense for me to provide such training, since they can then do it themselves.

I finally figured how to upload images... so here it is. in its full glory, 44 dongles, 22 on each hub

I usually do not appreciate art, but this looks like a beauty. the other beauty I appreciate are Pokemon trading cards.

internet has been down for a week... this is unusual. but something good came out of it. I bought a 5G router that is also a modem that I can insert a SIM card into. after testing I realised that actually that 5g modem/router is almost just as good as the fiber broadband I have. when fibre is down or undergoing maintenance, I switch the lan cable over to the 5g modem and no one can tell the difference. this seriously alleviates the problems the fiber optic is causing.



I also successfully set up a 2nd server on VPS. I think this is something that someone without coding knowledge can do. just follow instructions and ask for remote support when you need help.

a SIM card from rocket business (MyRepublic) has arrived. it shows m1 on the SIM card and im waiting to see if it disconnects just like the other m1s do. this is part of ongoing testing. I am always testing out new equipment, settings, software all the time to see where I can do better.

my fiber broadband internet is finally back up! apparently there are some configuration issues with the new static IP. the DNS has to be manually entered into each device. I had a hunch that it was DNS issues because the mobile proxies continued to work despite other websites not showing. this means the internet is going through but some how the device in the middle of it could not work with it.

so this static ip setting is supposed to reduce the cost required for the VPS server which costs like US30/month. static ip is free with my ISP provider. I plan to have 6 farms so that is 30 x 8 =240/month

I think this is just one of the many problems I should expect to face in this business. I have been forewarned by others that these problems never really go away and it's changing all the time. I think this applies to all businesses and somehow over here I am able to solve the problems and it does not frustrate me as much. I seem to have some kind of technical competence that allows me to fix things up.

the first mobile farm is fully sold out, and I am in the process of setting up my 2nd farm.

separately, I have been approached about providing mobile proxy services using a 5G phone. I think this is another offering that is part of the package if I want to be in the mobile proxy business. and also I will need to provide proxies from other countries. usually those who buy from one country would want to buy from the other countries too. I think this is called upselling or increasing customer lifetime value (CLTV). kind of like going to macdonalds, ordering a burger, a drink, fries, an apple pie and curry sauce condiments. now they charge 50 cents for each small tub of curry sauce. so the $4 burger becomes a $12 meal.

for the 5G phone, the phone costs about us100, compared to a dongle which costs us30. thats 3x the price. but the dongle is 4G, and 5G is the future. I saw that 5G also is more expensive, but only by 30%. using a real mobile phone seems to help the fingerprint/credentials more. but there is a problem with running multiple phones and that is battery bloat and I think that can be dangerous and explode. will I need to change the battery every 3 years? that would further increase the price of the phone. these S22+ galaxy Samsung costs around 300-400 when they are working normally. I usually purchase those with screen lines or errors and it's 1/3 of the price. the person using the proxy does not need the screen to be perfect.

I have researched into mobile data plans in Malaysia and it seems like all of them have a maximum of 1TB/month, which is the same as singapore. this means that I cannot use these cards as my base internet connection. some places do not have wifi and I need to use a mobile SIM card router to provide base internet. ideally, this should happen with a fiber optic broadband. the internet bandwidth on fiber is truly unlimited.

I am using a 3GBPS gaming fiber plan but when I looked around for business internet, they only offer up to 1GBPS and I wonder why. the pricing is also 3-4x more expensive. there is something I am not understanding well here. why are business plans charging 4x more while providing 4x less? thats 16x difference.

SIMBA works but with limits - their IPs aren’t rotatable, so the big 3 (Singtel, Starhub, M1) are still more reliable even if pricier. For downtime, regular restarts are normal with mobile proxies, but M1 going down more often might affect resale value. Using spare phones can work, but dongles are cheaper and easier to scale..

so recently I visited Malaysia Johor Bahru twice, to test out the connection for the possible mobile farm locations. the first obvious thing is that the reception in Malaysia is clearly much worse than singapore. in singaporean when we have 2-3 bars (4 being full bars) usually we will complain that there is a reception problem. but in Malaysia, having 1-2 bars is common and acceptable.

the location also affects which particular service performs better. in singapore usually there is not much of a difference and even if there is a difference usually our decision remains the same. but in Malaysia u really need to test it out to see which service performs better in that specific location. and also dont just depend on the bars. u should fire up a few different websites to see how fast it actually loads.

to order more than 10 lines as a foreigner, I will need to setup a business in Malaysia. initially the cost I found was more than us2500 and did not look feasible. however on further checks, there were budget setups that I could get for as little as US650.

it seems like my clients did not really need that much data monthly after all. checking from my admin panel, I can see most clients barely hit 20GB/month/line. so im not sure why they wanted so much data earlier on. my malaysian partner also does not allow me to use his fiber optic connection, but my buyer would accept having a 5G router instead of fiber. the bandwidth per month is only 1TB but my buyer accepted that for a farm of 30 lines. that means each line is only expected to use less than 30gb/month which is 1gb/day.

I have been thinking whether I want to set up in Malaysia or Thailand vs focusing on singapore. firstly there is not much to do in this business. so I can take up multiple projects at the same time. Malaysia is more costly than singaporean, and Thailand is more costly than Malaysia. this is because I am singaporean and there are alot of benefits and privileges and arrangements that a reduce cost for me. otherwise singaporean will probably be the most expensive place to setup.

I think of them as separate projects. there is a limit to how fast singapore business can grow while maintaining a certain profit margin or ROI. I can make it grow faster by investing more and accepting a lower ROI. I did that before and I feel that it affects my emotional well being and I rather earn less with a good margin and have more stable operations. its like Facebook ads where the maximum benefit is achieved through a low budget thats spread over as many days as possible. the first customer is always cheaper and better than the next one.

Malaysia revenue and profit is lower than in singapore, but the costs will also be slightly lower. if I was forced to choose in between Malaysia and singapore, it will clearly be singapore. but in this business I can have both. the natural expansion of singapore business will first be within singapore and the next best location will be Malaysia, so this is a expansion thats just waiting for the right time to happen. if thats the case, I would like to work on it earlier.

also I understand that proxy users would usually purchase more than 1 country, so the same clients that purchased my singapore proxy would also be interested in Malaysia or Thailand.

also, I understand that the regulations are looser in Malaysia, so whatever system I have created in singapore will work anywhere else. and I can relax those restrictions a little more and ease operational requirements.

this initial setup will cost usd650 and take about 2 weeks. equipment wise the investment will be about the same at us2000.

the static ip saves me US40/month/server. I intend to have maybe 7-8 servers so that will be around us300/month. apparently it also performs faster than a VPS. however in Malaysia I will not have a static ip, so I will have to purchase a VPS, which increases the cost of operations in Malaysia.

I also purchased 2 more 5G phones to test using them as proxies and also ordered more lines from ISP. unfortunately one of the ISP that was providing cheaper services have closed down.... so im left with the more expensive ISPs. the data allowance per month is only 150GB but from current operational experience it seems like its probably enough for all my clients.

xavierfok said:

Hi everyone, I recently started a mobile farm and would like to share my experience as well as interact with the BHW community.

I purchased a SIMBA card in singapore, which is our 4th biggest (and also smallest) telco and it seems like i im unable to rotate the ip when i click on the rotation link in proxy smart. it works for the other 3 telcos (singtel, starhub, m1) though. Does this mean that SIMBA sim cards cannot be used as a mobile proxy? or is it still useful but maybe certain functions are limited?

I also noticed that some of my m1 proxies seem to be going down more frequently than the rest. How serious is this problem? i heard that going down is normal for mobile proxies and they need to be restarted regularly. so how bad is it that it goes down more than normal? is the proxy still saleable and useful?

I'm happy to attach pictures but not sure how to do so. my setup is a optilex5050 with 16gb ram and sipolar 32 port usb hubs and huawei 3372h dongles

I'm also looking for information on business plans from singtel, starhub or m1 or their MVNOs. SIMBA is offering good business plans but the IP is not rotatable. As for the big 3, the advertised plans seems to be too expensive (5-10x normal user rate). but i noticed some MVNOs have this contact us for a chat function. do you have experience buying a data line as a business in singapore? i saw circles.life has one but im not sure whether they are going to offer it at a good rate.

---
I also noticed that there are real mobile phones available for rental at $250/month. How is that market different from mobile proxies? i have some spare phones as well, will they work or are they too slow? is it worth it to purchase a phone just for rental? its us200 here for a pixel 2, but only us20 for a dongle. the software cost per dongle is US1/month while the phone is US10/month. 10x difference here

Click to expand...

SIMBA cards may not support IP rotation if their network doesn’t allow dynamic NAT—so they’re limited for mobile proxies. Some downtime on M1 or other mobile proxies is normal; as long as the IP can reconnect, they’re still usable. MVNO/business plans can be cheaper but check if the data allows proxy use. Phones for rental mostly give you real device access, but dongles are much cheaper and simpler if you only need IP rotation—spare phones help, but cost/efficiency is lower than dongles.

liyong601c said:

SIMBA cards may not support IP rotation if their network doesn’t allow dynamic NAT—so they’re limited for mobile proxies. Some downtime on M1 or other mobile proxies is normal; as long as the IP can reconnect, they’re still usable. MVNO/business plans can be cheaper but check if the data allows proxy use. Phones for rental mostly give you real device access, but dongles are much cheaper and simpler if you only need IP rotation—spare phones help, but cost/efficiency is lower than dongles.

Click to expand...

SIMBA cards are much cheaper than the rest, I think they actually own some infrastructure. not sure why they dont allow IP rotation though. I think downtime is fine, as long as it can reconnect like you mentioned. I think sometimes it doesnt so I have trigger a manual restart. am I doing this correctly? or is there a setting in proxysmart software that can auto-reboot? I tried playing with the settings but it doesnt seem to be working the way I want it.

I found a software on 5G phones called localtonet and it does not give access to the mobile app or sms or calls, just a tunnel and apparently thats all you need. I dont want to allow SMS or calls if possible because the regulations in this area is strict and place alot of liability on the owner of the number.

if I could choose, I prefer dongles. but I dont get to choose everything in life. I think it's a separate market. while the cost/efficiency is lower, I think this 5G phones market on its own makes a valid business case. I dont think the dongles are an infinite scale. I think it's limited. so it will come a point where it does not make sense for me to invest in any dongles in singapore anymore. u are judged according to the IPs around you, and if they are all proxies, you are dead meat. so we need at least 99% of them to be normal organic traffic. data analysis also requires a certain sample size. if its too big, it will get picked up by checking mechanisms.

I just learnt recently that u-mobile from Malaysia does not work on the e3372h dongle. further investigation indicates that the ISPs block use of mobile SIM cards on dongles. they want people to use broadband SIM cards on dongles and 5G routers. I think this is a small problem. they just want you to pay more, not to actually stop or limit any of the operations. so it could happen anytime that one of the ISPs block dongles permanently and I will be a sitting duck with all these equipment.

previously I was only slightly inclined to start operations in Malaysia, but based on what I have seen, im now much more encouraged. firstly I dont think its so easy to start a farm without any experience. I have someone that is willing to teach me. usually I should be willing to pay for this. but in this case, im selling the proxies at a big discount, which is actually a great deal. I dont expect profits from this setup deal, but only for the new customers that I bring in at market price. Malaysia regulations are also more relaxed than singapore in general. if any one country stops or restricts operations, I can shift equipment to the other one.

I went to Malaysia 2-3 times to scout for possible locations to place the farm. unfortunately the stars dont align. those that had the right ingredients were too expensive. and those that were cheap enough did not work or was way too far away. I guess this project will have to be delayed until I get critical volume and can take up a more expensive location with maybe 100 lines right at the start. this is a stark difference compared to singapore. maybe because I am singaporean, so many of the requirements to set up a company can either be filled by myself, or more relaxed, or faster.

I have learnt that a 100mbps fiber broadband connection can support 4 servers which is around 160 proxies. I can have a maximum of 200 proxies per ISP, and there are 3 main ISP in singaporean so that will work out to 600 proxies, which is 15 servers per physical location. so the minimum fiber broadband connection I need is less than 500mbps. I noticed that the business broadband plans in singaporean caps out at 1GBPS. my current home fiber connection is at 3GBPS which is way more than enough. its funny though, there is another 10GBPS connection but all these higher plans are not available to businesses. also in Malaysia, the business plans are also worse than the consumer plans in general. maybe its the competitive landscape that made the individual plans much more attractive.

most of the m300z and e3372h dongles on aliexpress have been purchased. the rest are either not unlocked or at a price higher than Alibaba. so I have purchased my first bulk order of 50 m300z at US15/each. I previously paid up to US30 for a e3372h. I also made some other test purchases for e3372h which I can get at us20/each.
I also purchased a gigabit 8 port from netgear to attach to my modem to allow 7 more servers. I like to purchase equipment, it's like a toy.

I have gotten in touch with m1 ISP provider and it seems like I can order bulk purchases of more than 40 at a go. and its a 3month contract which is worse than starhub but I have to take whats on the plate for now. singtel is even worse, they want a 1 year contract and it seems like they are not responding to any of my emails.

I think that the people I meet at BHW should be seen as a long term relationship rather than an instant sale. they seem to be looking around and testing and if I chat with them regularly and see what they need and offer help, they are more likely to give my service a test and see if it works for them. they are also in a testing phase and figuring out what would work for the them and when there are so many offers they can get overwhelmed and not consider what I have.

one of the electrical plugs for the USB hub came loose and I have to put it back in position. its still loose so I had to put a cable tie to hold it into place. I think there will be regular problems like this maybe once a week that requires physical attention. I can only do this in singapore. I have 2 server locations confirmed and that's 1200 mobile proxies so I think that is enough for now and I wont have to consider more server locations unless there are buyer requests.

I noticed that so far no one really uses more than 50GB per month but when I offered 150gb, it seems like some of them wanted more, up to 300gb or unlimited. I also purchase more data for myself but never actually used it. there is a maximum physical limit to how much data can flow in each proxy and usually they dont need the speed. they need the latency to be low, and singapore proxies are poised just to do that. I have stable fiber connection, 4G/5G network, stable electricity, custom routing for both fiber and 5G connections to offer the best proxy experience ever. best of all, none is something I had to work on, its all given by the government.

so it turns out that the 5G router I had as a backup to fiber broadband has to be configured and installed separately and physically. I could not setup an automatic failsafe. basically I have to redo the setup again as if I never had fiber broadband. so I placed it in a separate internet network so that they could control each other in case one of them went down. I guess this is the extra cost that comes with providing reliable proxies. additional router cost, subscription, modem, cables etc which I estimate is maybe only 5% of the ongoing variable costs.

it seems like the best way to grow so far is to have a BST in the proxies section and follow up with it everyday. so instead of putting all my energy here, I should give it time to grow slowly, and place 95% of my energy somewhere else for maybe 2 years before this project gets big enough to demand more attention. I'm looking for a job in marketing, and also taking up a course in data science to switch an industry to something more in demand and less like a commodity.

Just got a new rack to put everything together in a more organised way.



Recently I applied to study data analytics to change my career to become a data scientist. Lessons I learned before and how I will study now - read everything, interact, research, do quiz and then read everything again and then discuss with everyone else and do your project. And if you have spare time, repeat the process. Specially for exams, keep doing the same questions to commit to memory. For projects refer to past projects mostly. But thats for grades. This one im going for is a pass/fail. So I guess I will use my time along the way to build something I care about, and increase the chances I will pass. And along the way learn newer and more things that I couldn’t if I was completely new.


Life organisation - I work best when busy all the time, right now it looks like the main activities will be
Data science/analysis job
Mobile proxy services
Magic the gathering card game

I have found new telcos who are able to offer more data for lower prices. I think this is something thats always changing and I will need to keep myself updated regularly, as with all aspects of business and life.

as for the physical locations, it seems like something different pops up every week and requires me to be physically there to fix it. new software, new hardware, new configurations, new phone etc. new locations also cost money and the setup looks like a bitcoin farm and I get complaints about it. I also need the fiber optic and a lan cable, adding more to the sore eye.

It’s been about 5 weeks into the data analysis course. I can’t see exactly how this course can map into the mobile proxy I’m working on, but I believe that it will cross roads in the future. Data analysis is like math or coding, required everywhere but I haven’t figured out where.

I created an app that tracks laptop prices on eBay and helps resellers decide on the price to buy and sell. I posted it on Reddit and it gathered some responses and it shows that there is some traction. I spoke to my instructor and it seems like I have to code this app all on my own, and not with ChatGPT like I did. So I’m going to do this all over again from 0. The MVP is already out and it’s proven it can be done. But I’m not sure if I’m good enough for it. But that’s what life is about. Doing something further than you can reach.

So I have been paying for “backups” to ensure that the mobile proxy does not go down. Got a 5G modem router for 200. It’s great and can support 2 full servers on its own. But for the last 3 months it’s just there collecting dust.

I also kept a few extra subscriptions in case something went down. That eats into the cost.

Every now and then I get reminded of why this is necessary. For example, I intentionally split the wifi electricity control to be in different router network so that I can always use one to control the other. The alternative is that one controls everything and if I switched it off, it’s not possible to switch it back on wirelessly and I will need to physically trigger the switch.

I accidentally turn on this switch occasionally but any one occurrence pays for itself.

I just had a “win” that made my brain louder.

I bought a Riftbound Origins booster box at US$1,000.
I flipped it for around US$1,700.

On paper? Clean. +70%.

But the moment I sold it, a more annoying question showed up:

Did I actually maximize the value or did I just get lucky that it sold?

Because when you watch eBay long enough, you start seeing the truth:

There are tons of listings at crazy high prices, that never sell.

A lot of those “market prices” are just wishful thinking with a Buy It Now button.

And the real action happens in offers, buyers casually asking for 10–20% below list like it’s normal.

So you realize something:
People don’t get paid from listing prices.
They get paid from transaction prices.

Not “what sellers hope for.”
But what buyers actually check out at.

That’s why this US$700 profit feels good, but also suspicious:

Could I have bought cheaper with patience and better timing?

Could I have sold higher if I waited?

Or would I have waited too long and gotten trapped in “display prices” with no real demand?

And here’s the uncomfortable truth:

The market isn’t about who can list the highest.
It’s about who understands liquidity.

How fast it moves.
At what price it actually clears.

Because “selling high” is cute.
Selling consistently is the real skill.

Most people confuse:
listed price with real price.

And that mistake is expensive.

So I have just completed my data analytics course from General Assembly. The last month was really busy working on the projects and homework. I hope these skills will be helpful to me at work and also in my personal life.

I have listed my proxies on several marketplaces and maybe about 15 of them are active right now. I cant find any more marketplaces but I think that I will need my own way of getting traffic so I can become self reliant and each unit of effort I put in will bring in marginal gains and hopefully compound effect as well.


I heard alot about Claude lately and I purchased to see whats the big hooha. Turns out, it saved my life when one of my servers crashed. Apparently there were alot of log files because some clients were rotating IPs ever year minute. But the initial errors would not inform me of that. It said mongoldb error, restart it. And then it said usb devices not connected. But after 40-50 cycles of taking picture and uploading on Claude, the problem was finally resolved. It took about 30 mins and my clients didnt look for me within that time. I think this is magical.




Also I created dashboards, google sheets and web app to help me to manage my proxies. It’s far from done, but the initial setup is so promising. It actually works and I have the files saved systematically so I can continue working on them on a daily basis.


I also made use of it to write SEO articles and it brought 120 impressions per day from 25/day. Still no clicks yet but I think if I can spot a new interest, I would be able to get some clicks. These older keywords are highly competitive and I wouldn’t be able to rank even if I can improve my article significantly.


All these were done within 2 weeks.


As for the trading cards before, I have sold off most of it and I will be selling the main prized possessions as well. I will be moving my interest into coding/claude/online business instead. It took a while for me to find the right group though. The trading cards were a source of emotional support while I discovered my new interests.

So now have 4 servers running… 1 for each dedicated client, one open claw server that I cant figure out how to use. After going thru so much coding, I started to notice that some commands were common, like bash, grep and the —.


As for marketing, I intend to focus on SEO, podcasts and use that content to make into videos. I have already created 10 YouTube videos and podcasts that are scheduled. As well as about 90 pages of SEO. Let’s see if I can find that sweet spot that recently came up and others ignored and im the first.