Get live statistics and analysis of ildar's profile on X / Twitter

I tested multiple VPS providers for my projects. Here are the results I needed a cheap and simple solution to start with, because I wanted to create a lot of small projects to test the market. I tried: - Aeza, PQH (THE HOSTING now) - some of their servers crashed and were down for a while so I decided to try different solutions - Digital Ocean, AWS, Google Cloud - cheap at the start (from 4$) but scaling is expensive - Hetzner - asked me for my documents and I didn't proceed - OVH - cheap but I wasted 2 hours trying to connect to the server via ssh, so I got frustrated and asked for a refund - Webdock - cheap (from 2.5$) and convenient. The scaling price is very good. It's the one I'm using right now. I have just 1 concern: there's only 1 location and if it goes down (like a power outage or internet issues in the area) it will affect all my systems. But I'm taking the risk for now and I saw they plan to add more locations I know people also use serverless architecture for SaaS. I've never worked with it. Is it a good choice for new projects? I just use docker for my apps on a vps

Do SaaS customers tend to prefer Stripe, or is it ok to use an alternative? What alternative would you recommend?

If you're a programmer and you think about building algo trading bots here's my 3 years of experience failing at it At first I tried using ready made solutions like DCA bots, other bots on Binance, copytrading and investment platforms that claim to have profitable algorithms. I'd come home after work hoping to discover gold that would help me build some passive income and potentially leave 9 to 5 job. But eventually all of them failed and I lost the money (~800$ I think). Then I discovered there's a thing called Crypto Flash Loan Arbitrage. The idea is to buy and sell crypto in 1 transaction using money from a provider. The money is returned in the same transaction, so the bot doesn't even need equity (except for the transaction fees). I even learned Rust to develop a super fast bot. But when I ran the tests, there were no real arbitrage opportunities. And I thought even if they existed they'd likely be in low liquidity/new tokens so I decided it wouldn't be worth it. Later I learned there are bots that do this arbitrage before the transactions are even recorded to the blockchain. It's probably those bots that take all the opportunities. And the effort to build one was just something I wouldn't take. So I took a break. Some time later I decided to try again but with a live trader who knew real strategies. There was 1 telegram channel with trading signals that stood out because the signals seemed real and profitable. I checked the charts and it really looked like it'd work. But when I actually built the bot it turned out the signals were coming with the delays so long (1-20 mins) that we only lost money Then came the period when we were trying to replicate the algorithm from that channel and build our own strategies. We failed We even tried automating the trader's live strategy, but it also didn't work. It was profitable when he traded manually but failed when the bots were doing it. So, I guess human intuition plays a huge role in this game. The last thing I tried was a crypto futures funding bot. Interestingly, whenever the bot went long, it lost money and when I switched it to short, it lost again. It felt like the exchange's algorithms were sorting API orders to disadvantage them. So I never made any money and lost a ton of time. Is it even possible to make money with algo trading? I don't know. I don't want to do it anymore. Although, maybe it works for big firms with teams of quants with phd. Maybe for a few lucky individuals who discover oil in the desert. Maybe it works for skilled traders who know the market very well and just need to automate their strategies. If you know anyone who actually profits from algo trading, please comment I'm curious. But for now, my opinion is that 95% (or more) of bots, algos and products with them are either failures or scams. What do you think? Is this whole industry just a trap or it was my lack of knowledge that led to the failures?

I like that I can work on my X while AI is working on my code

I personally try to avoid subscriptions because I'm afraid I'll forget about them for months. And I don't want to manage them So why does everyone build SaaS with a subscription model? Is it better than pay as you go?

As a customer, would you like to have the option to pay for SaaS with crypto?

I used to check the traffic of similar apps with similarweb, but then I switched to semrush because it doesn’t require a subscription. How do you analyze the traffic?

If you need to automate a simple task but you don't want to deal with servers there's a thing called Google Apps Scripts (GAS) It's absolutely free and it's basically js scripts you can connect with google sheets, docs, slides, gmail and so on One example is if you have a google sheet with messages and emails you can send them automatically with GAS from you email

I'd love to use them! - translation - recalling something - calling - maps - asking for advice - smart home management - health/sport metrics in live like heart rate, steps - working assistant But... Ads right in the field of view!! Will they dare to do it?

Similar idea: an alarm clock that charges every time you press snooze

Startup idea: keys for cpuchain

Some night coding

Please help me with one of my projects. I need the community's opinion on what to do next Project description The project is made for Telegram. It's a bot in which users can connect custom sources like RSS, X, Telegram channels and chats. The bot reads all new posts from the sources and sends them to LLMs with the user's custom prompt. This way, users can filter posts from their sources and generate new ones. Sources └── [posts + custom prompts] └─> LLM ├─> discards post ❌ └─> sends to User ✅ Possible applications include: 1. Fully or partially automated Telegram channels with news from RSS, X or other sources 2. Custom post filter, so users can gather messages from different sources in 1 channel Issue description I found it really hard to sell, and it's very difficult to explain its value. This idea isn't really unique, but I couldn't find real competitors. So I can't even learn from others or figure out where my customers are. On top of that, most paid Telegram bots are made for Russians, but with Telegram getting restricted there, I’m not sure the global market exists. And I'm trying to move away from doing business with Russia. Besides, I have concerns about how people will use it. I don't mind if users create fully automated channels as long as they're open about it. But some might lie about it and not even check the posts made by LLMs. So the question is: what to do with it? I could try to sell it for its technology, since the project's revenue is very low. There is another option that interests me and I would like to know your opinion about it. It's to adapt this technology to a different demand. It wouldn't be hard to migrate this project from Telegram to X. But my concerns about its usage on X are even stronger. Do you think this project has potential in another related field? If so, which one?

Is there an app that can analyze current trends and demand in SaaS so I can choose what to build next?

Opposite idea: AI ad unblocker

Do you listen to music while coding?

If you're a programmer and you think about building algo trading bots here's my 3 years of experience failing at it At first I tried using ready made solutions like DCA bots, other bots on Binance, copytrading and investment platforms that claim to have profitable algorithms. I'd come home after work hoping to discover gold that would help me build some passive income and potentially leave 9 to 5 job. But eventually all of them failed and I lost the money (~800$ I think). Then I discovered there's a thing called Crypto Flash Loan Arbitrage. The idea is to buy and sell crypto in 1 transaction using money from a provider. The money is returned in the same transaction, so the bot doesn't even need equity (except for the transaction fees). I even learned Rust to develop a super fast bot. But when I ran the tests, there were no real arbitrage opportunities. And I thought even if they existed they'd likely be in low liquidity/new tokens so I decided it wouldn't be worth it. Later I learned there are bots that do this arbitrage before the transactions are even recorded to the blockchain. It's probably those bots that take all the opportunities. And the effort to build one was just something I wouldn't take. So I took a break. Some time later I decided to try again but with a live trader who knew real strategies. There was 1 telegram channel with trading signals that stood out because the signals seemed real and profitable. I checked the charts and it really looked like it'd work. But when I actually built the bot it turned out the signals were coming with the delays so long (1-20 mins) that we only lost money Then came the period when we were trying to replicate the algorithm from that channel and build our own strategies. We failed We even tried automating the trader's live strategy, but it also didn't work. It was profitable when he traded manually but failed when the bots were doing it. So, I guess human intuition plays a huge role in this game. The last thing I tried was a crypto futures funding bot. Interestingly, whenever the bot went long, it lost money and when I switched it to short, it lost again. It felt like the exchange's algorithms were sorting API orders to disadvantage them. So I never made any money and lost a ton of time. Is it even possible to make money with algo trading? I don't know. I don't want to do it anymore. Although, maybe it works for big firms with teams of quants with phd. Maybe for a few lucky individuals who discover oil in the desert. Maybe it works for skilled traders who know the market very well and just need to automate their strategies. If you know anyone who actually profits from algo trading, please comment I'm curious. But for now, my opinion is that 95% (or more) of bots, algos and products with them are either failures or scams. What do you think? Is this whole industry just a trap or it was my lack of knowledge that led to the failures?

I personally try to avoid subscriptions because I'm afraid I'll forget about them for months. And I don't want to manage them So why does everyone build SaaS with a subscription model? Is it better than pay as you go?

Some night coding

Do SaaS customers tend to prefer Stripe, or is it ok to use an alternative? What alternative would you recommend?

I like that I can work on my X while AI is working on my code

Similar idea: an alarm clock that charges every time you press snooze

Please help me with one of my projects. I need the community's opinion on what to do next Project description The project is made for Telegram. It's a bot in which users can connect custom sources like RSS, X, Telegram channels and chats. The bot reads all new posts from the sources and sends them to LLMs with the user's custom prompt. This way, users can filter posts from their sources and generate new ones. Sources └── [posts + custom prompts] └─> LLM ├─> discards post ❌ └─> sends to User ✅ Possible applications include: 1. Fully or partially automated Telegram channels with news from RSS, X or other sources 2. Custom post filter, so users can gather messages from different sources in 1 channel Issue description I found it really hard to sell, and it's very difficult to explain its value. This idea isn't really unique, but I couldn't find real competitors. So I can't even learn from others or figure out where my customers are. On top of that, most paid Telegram bots are made for Russians, but with Telegram getting restricted there, I’m not sure the global market exists. And I'm trying to move away from doing business with Russia. Besides, I have concerns about how people will use it. I don't mind if users create fully automated channels as long as they're open about it. But some might lie about it and not even check the posts made by LLMs. So the question is: what to do with it? I could try to sell it for its technology, since the project's revenue is very low. There is another option that interests me and I would like to know your opinion about it. It's to adapt this technology to a different demand. It wouldn't be hard to migrate this project from Telegram to X. But my concerns about its usage on X are even stronger. Do you think this project has potential in another related field? If so, which one?

Will this end all the debates about which browser is better?

As a customer, would you like to have the option to pay for SaaS with crypto?

I'd love to use them! - translation - recalling something - calling - maps - asking for advice - smart home management - health/sport metrics in live like heart rate, steps - working assistant But... Ads right in the field of view!! Will they dare to do it?

Startup idea: keys for cpuchain

I used to check the traffic of similar apps with similarweb, but then I switched to semrush because it doesn’t require a subscription. How do you analyze the traffic?

I just started reading this guy's 2nd book. I'm on 11th page and I found it incredibly relatable

Is there an app that can analyze current trends and demand in SaaS so I can choose what to build next?

If you need to automate a simple task but you don't want to deal with servers there's a thing called Google Apps Scripts (GAS) It's absolutely free and it's basically js scripts you can connect with google sheets, docs, slides, gmail and so on One example is if you have a google sheet with messages and emails you can send them automatically with GAS from you email

I tested multiple VPS providers for my projects. Here are the results I needed a cheap and simple solution to start with, because I wanted to create a lot of small projects to test the market. I tried: - Aeza, PQH (THE HOSTING now) - some of their servers crashed and were down for a while so I decided to try different solutions - Digital Ocean, AWS, Google Cloud - cheap at the start (from 4$) but scaling is expensive - Hetzner - asked me for my documents and I didn't proceed - OVH - cheap but I wasted 2 hours trying to connect to the server via ssh, so I got frustrated and asked for a refund - Webdock - cheap (from 2.5$) and convenient. The scaling price is very good. It's the one I'm using right now. I have just 1 concern: there's only 1 location and if it goes down (like a power outage or internet issues in the area) it will affect all my systems. But I'm taking the risk for now and I saw they plan to add more locations I know people also use serverless architecture for SaaS. I've never worked with it. Is it a good choice for new projects? I just use docker for my apps on a vps