Building a Bot Army

Can a guy be in love with his basement walls?  OK, I’m being hyperbolic now, but seriously my basement wall gpu miner has been running for 13 days straight without so much as an increase in the GPU temps.  This means that I use less electricity to spin the fans!  This is working and I can’t wait for the electrician to come back from vacation so I can hang my BTC miners!

Step two: Node cluster.

I have several nodes that I run – these are discussed by me in other areas of this website, but for now they need a home.  It’s not quite as exciting to talk about this, but it’s needed for me to succeed.

I have overseen the creation and maintenance of some pretty complex automated clusters of servers.  All working together to serve internet applications and heal themselves as they run into hardware failures.  I could build this, but I decided to rely on a third party to build and support it.

I gave them the specs and set them free.  Eventually they’ll get back with a plan and I don’t have to worry about it.  Outsourcing this step leads me back to my bots.

Step Three: My BOT Army.

I haven’t mentioned much about my bots in the previous blogs because I have adopted a kind of strategy to keep up with them.  Every morning I wake up and go right to the bots – restarting any that need it.  Every night I go through and look for better bots and replace the bots that look to be not producing well with others that show better promise.

This works kinda…  It certainly gets results as I make about 22% off my Bitsgap bots monthly, the only problem is that I don’t have much funds in each bot, so 22% of not much is a real little bit.  Certainly not something I could live on.

I have been exploring with non-bitsgap bots too – kucoin has bots on their ios app, 3commas has bots, and so too does pionex.  I want more bots with bigger balances.  If I can, I’d like to have as many of my assets in a bot as that’s what I see them there for.

My intention in 2020 was to buy a little of a lot of crypto – amassing a large range of assets.  I wish I bought more and was more selective, but I’m happy with the results since I’ve already de-risked my entire portfolio.   This means that I’ve earned back all of the USD that I initially invested.  I plan on investing more USD at some point but for now I can grow the crypto with my bots and buy the dip for any missing assets.  Though, once the market takes off again, it may be difficult to find a decent entry.

My Theory

I believe it’s a relatively safe investment to open a quoteless bot.  I have taken a full year to test this thesis and in most all cases it has proven to be true.  I go into this in greater detail here, but no matter the strategy I need to build a proper process to maximize the gains.

My goal is to have as many bots as possible running in the green without spending all day monitoring them.  It seems a lofty goal, but I believe it’s something I can achieve with a solid routine and my iPad.

A bot is a computer program using my crypto assets to buy low and sell high.  Grid bots make a grid of potential trades and once one trade is executed the make profit trade is placed.  Each round trip generates a tiny bit of crypto and each bot has the capacity to make several thousand round trips daily.  The challenge isn’t the bot mechanism, it’s rather watching the market effectively and starting bots when the market moves up and on the right pair to make a gain.

This requires a bit of a re-focus of my efforts and a utilization of the private analytical environment I built.  I have set up rudimentary data flows and built a few dashboards with information that I can use to start and stop bots.  The idea is to use the information to stop the bots that aren’t being as profitable and start new bots that look like they will be really profitable.

BOT Monitoring  

The existing bots are spread among several platforms and I currently monitor them manually and it sucks…. I have 30 or so that I check into each day at least twice and the bulk of my time is taken up with the mechanics of setting up and breaking down bots.  The metrics to monitor are all buried inside random companies’ apps and websites.

I have tried rather unsuccessfully to get a proper data feed from bitsgap, so I have to rely on logging into their site, copying and pasting, reformatting, and re-calculating things in order to understand which bots are working better.

I’m also limited by what I can gain – currently I see only bots gaining USDT, BTC, or ETH as those are typically the most common quote currencies.  But I want to build a free base in COMP, MKR, and other assets like this…. I absolutely need to re-think this bot concept as it’s proven to be semi-effective, but there are all of these monthly fees and such that I need to pay.  3commas, bitsgap, etc…. I certainly make more than the fees – but having a bot-rich, feeless environment is what I’m truly after.

I want to build infrastructure in my basement datacenter in order to automate the entire process from buying to selling at a profit or even selling and buying back for much less…. This is where I’m headed, but first I have to organize what I have and establish a proper starting point.  The tools I use now are easy to use – lets see if I can push their profitability through the roof.

Actual Profits

If I want to live on what I earn I need to be carefully focussed on profits of course.  In a traditional job one receives a paycheck and deposits USD into their bank account.  They then use this USD to buy groceries, necessities, things they want and some hopefully goes to save and invest.

I don’t get USD at all but I still need to spend USD to live.  So how can I do this?  How can I earn crypto and spend USD?  Easily, as nearly all crypto has an indirect path to USD and can be “cashed in” anytime I want.  This presents me with opportunities to make gains just by holding crypto and selling at the right time.  It’s the same concept as the stock market where every stock has a USD value that evolves over time.

Introducing the “stable coin”

There are many stable coins in crypto, including USDC, USDT, TUSD, etc…. Each of these stable coins are pegged to the value of the USD with some sort of automated mechanism.  This means that 1 USDC = 1 USDT = 1 TUSD = 1 USD.

We can measure our assets by how much they are worth in stable coins.  Over time I trade crypto for stable coins to lock in profits just the same as you’d sell Amazon stock to lock in USD profits.

I know that I will need several bots focussed on earning USDT, but I can also have other bots focussed on earning BTC or ETH as those assets appreciate over time and I want to sell them when it makes sense and/or I need the USD.

So say I have earned 0.001 BTC today – that’s about $38 in today’s market.  I could trade that for USDT today and load my crypto.com credit card with that USDT.  Then I could pay for something with the credit card and it subtracts from the $38.  The only catch is that I need to earn 30% more for the taxes…. So the $38 is really only $26.60 as the remaining $11.40 is the tax that I must pay for the $38 gain I received.

It’s actually quite costly to be in crypto as crypto is now seen as a capital gain and the taxes are near 30%.  In a paycheck the income tax comes out before you receive the money so it’s easier to plan for taxes as you already have paid them by year’s end.  With crypto you can certainly get caught if you didn’t plan ahead.  If I’m going to live like this, I have to plan for taxes.

A lot of thought needs to be put into what I mine and when I sell what I mine.  This is where my experiences with CryptNation come into play.  I am NOT a trader, but I am able to read a chart.  If I can, I’ll position sales of assets I mine at a gain rather than sell daily (like some miners do automatically.)

Take the miner now making .19 ETC on average per day in my new datacenter.  1 ETC is currently worth $48 but last night 1 ETC was worth $52. That puts yesterday’s catch worth $9.88 last night but if I were to sell right now I’d only get $9.12.  This is an economy of scale thing as even $.50 less per day adds up over time.  At ETC’s height that .19 ETC per day would have been worth $30.40, so there is some chart reading needed to set up proper sell orders.

The goal is to mine optimally and sell the mined assets to stable coins at the highest value so I can reap the greatest rewards.  I will hold my gains in USDT and sell the USDT for USD when I need cash to pay a bill.  I can use these newly mined assets to start bots in various systems as I’m positioning them to sell high.

As I look forward to this next week I am excited about getting the electrician to come and run power so I can finish my basement mining datacenter.  I also face a challenge to build an easy to understand dashboard with all the info I need to organize and better monetize my bot army.

And I must take a high-level look at my portfolio as the market has begun to rise again.  I need to set out some sell orders and rotate some of the assets I have owned the longest as they begin to come back into the green.  I need to get bigger stashes of the better BOT performing crypto assets.  The only real guiding principle is buy low and sell high – the rest is more art than science.