Getting the transcript
Reading the captions from YouTube. A video nobody has opened here before takes 10 to 30 seconds; this page fills in on its own.
Getting the transcript
Reading the captions from YouTube. A video nobody has opened here before takes 10 to 30 seconds; this page fills in on its own.

Trading with DaviddTech · @DaviddTech
This video has no Most replayed graph yet: YouTube shows one only once a video has enough views. These are the moments viewers replayed most in Trading with DaviddTech's most watched videos.
Most replayed moment at 9:16
4.8x that video's typical replay level
installation of this is pretty simple. You can use Python or something like that, but I have made you this absolutely wonderful, fully detailed workbook, so you can go through building your own dashboard and installing it just by prompting Claude.
Said at 9:09
The graph counts replays. It does not show where viewers stopped watching.
Words
4,103
Runtime
22:38
Speaking pace
181wpm
Reading time
17min
181 words per minute, the same as the 181 median of 349 measured videos. That distribution comes from the 349-video hook study.
Opening (first 30 seconds)
In this video, I'm going to show you exactly how you can build your own AI trading bot. Because when I started trading, I lost money twice. Once to manual trading and once to running somebody else's bot. Then I spent over 5 years back testing strategies until I became finally profitable. So, if I had to start again today with only $1,000, this is the exact bot build that I would do. It's the exact same kind of bots that I used last year to double my account. We are
91 words, the words spoken in the first 30 seconds at 181 words per minute.
Free, no signup. See how the first 30 seconds hold attention, with rewrites.
Sentence shape
| Measure | This transcript |
|---|---|
| Sentences | 255 |
| Average words per sentence | 16.1 |
| Longest sentence | 86 words |
| Questions asked | 3 |
| Sentences containing a number | 25 |
Most used terms
Filler phrases
47 in total: actually 22 · like 21 · literally 2 · basically 1 · kind of 1.
A literal whole-word count of the same phrase list the Prepublish browser extension uses, so a phrase inside another word is not counted and a phrase used in its ordinary sense still is. It is a count and not a judgement.
Free, no account. See where attention is likely to drop, with a rewrite for each weak line. The free check shows the scores and the one issue costing the most. Or run it on the words above first.
Free · No login · See a sample audit first if you prefer.
What this transcript is
Every word below is the caption track YouTube publishes for this video, pulled from the video itself and reproduced unchanged. It is not Prepublish's writing, not a summary, and not a re-transcription: it is the video's own published captions. English captions, generated automatically by YouTube, in the video’s original language. Source: the video on YouTube. A channel that would rather this page did not exist can ask for its removal through the contact page, and it is removed.
In this video, I'm going to show you exactly how you can build your own AI trading bot. Because when I started trading, I lost money twice. Once to manual trading and once to running somebody else's bot. Then I spent over 5 years back testing strategies until I became finally profitable. So, if I had to start again today with only $1,000, this is the exact bot build that I would do. It's the exact same kind of bots that I used last year to double my account.
We are going to be building it step by step. The great thing is that you will not need to know how to code. We're just going to be talking to Claude or Jack GBT. It's free and I'll show you the full process so that the end you can steal it and make it your own. But here's the thing that nobody tells you. One bot is not enough. I will show you why and exactly how many bots $1,000 can actually run by the end of this video.
But for all of that to happen, we're going to have to start right now. Okay, before we get over to my computer, let's have a look at today's road map. First of all, I'm going to be showing you the tools that we're going to be using and their real costs where possible. I'm going to be using things that are free of charge or you most probably already have their subscription. Next, I'm going to show you how you can generate your first trading idea.
I will include a couple of rules that you'll have to take into account when only trading $1,000. We're going to build and back test the idea because you don't want to be risking your own money on something that you don't understand. Then, I'll talk a little bit about incubation, which is probably one of the most important steps. Finally, we're going to connect that bot up to your real broker and trade $1,000. At the end, I will go through a bit of real talk and give you five rules that I use every single day in my real life trading.
And guys, if you are following along and building your own AI bots today, if you don't mind giving a good old thumbs up to this video, it really does help. And why not subscribe? I do videos every single week about AI trading. You can find more links about what we're building at school down below in the description. Right, let's get on with our first bot of our profitable AI agent. Okay, here I am on my computer. Let's get this project started.
Okay, first thing we're going to need is to choose a AI. I personally like Claude myself and very much Chat JPT as well, but in today's example, I'm going to be using Claude AI. You can use any model to do this. We're not going to be stuck inside the Clawude ecosystem. It really is just a matter of taste. Personally, I'm going to be using Claude. So, I'm going to click here and download Claude MacOSS. Now, obviously, Claude comes with a price ranging from free all the way up to $90 per month.
Now, with Chat GBT, you can actually use the free version to use Codeex where we're going to be coding our bot. I personally have the max version because I use it day in and day out. Next, we're going to need a Trading View account. Now, the reason we're going to be needing a Trading View account is because Trading View is one of the most cost effective ways of actually getting data easily back testing strategies and also connecting them easily over to your broker.
Now, once again, you don't have to use Trading View. You can go away and download all of the data from your broker, but for me, I found that Trading View just simply is the most cost effective way of doing this. Now, they do have a free plan that allows you to just set up two alerts. We will be needing those alerts and two will be enough to get you started. Next, we will have to give our AI agent a set of tools. Those tools will allow it to back test strategies, build strategies, and also live trade them.
I'm going to be using tradingkit.com. The reason is is it's free and is a wonderful community project. They've backed it over 1.1 million strategies which you can find directly on the website and you can easily import those strategies over to your own trading view. So I'm going to be using tradingkit.com. And finally, we're going to need a middleman service. This middleman service sits between your broker and your AI agent and allows us to take live trades.
I'm going to be using trigger.trade because it is anonymous and it is free of charge. But you can use any of these middleman services. And there are multiple brokers already integrated into Trading View at no extra cost as well. Okay, we're ready to start building. One thing that you will need is a broker account, but we will tap onto that at the end of this video. Right, let's start by downloading and installing Claude code.
Okay, as I said earlier, you can use Chat GBT if you want or any other AI model. I'm using Claude, so I'm going to download the Mac version just here. Clicking here, downloading it to my desktop. As you can see, I already have it installed and downloaded. I'm going to go over to the sidebar here. I'm going to be using Claude code and not Claude chat today. Although, I do like Claude Cowwork for exactly this model as well.
Okay, so we have Claude up and ready for us to start working today. So, the first thing that we're going to do is we're going to go over back over to tradingkit.com. We're going to log in with a free account. I'm already logged in here and we're going to press my account. Now, just as a reminder, this is just giving tools to our clawed AI agent so that it can actually back test our strategies. This is going to stop us from losing money and helping us to back test our ideas easily.
So, we're going to go down here. You can see the code just here. We just copy that over to Claude, paste it into the box just here, hitting enter, and that will go away and install. It should take around 3 minutes to install. I will not install it today because I've already got it installed on my computer. Next, it will ask us to go over and pick up this authentication code. We're going to copy that, paste it into the box down below, and hit enter once again.
That way, you are authorized to actually back test strategies on that website. So, the next step is just literally to make sure that there is a connection between trading.com and Claude. To do that, we're just going to say, "Hey there, buddy. Could you back test a very simple trading strategy for me? Do it on the 1 hour time frame. I just want to make sure that the connection between you and my MCP server is working perfectly.
For this, I'm going to be using Opus 5.5. I'm going to move it down just to make it a little bit faster to medium. We're going to hit enter, and that's going to go away and do a simple strategy." And boom, there you have it. back tested a simple strategy over a one-year period. No, neither here nor there. Doesn't really make any difference. These are the trades. Here's the win rate, profit factor, and all of this has built a Pine Script strategy, which you'll be able to import over to Trading View later.
So, once that's done, we have done probably one of the harder steps of actually building these strategies, connecting these two websites together. Next, we're going to have to come up with some ideas of strategies that potentially will work. So, here is where I go away and find my strategies. First of all, I go to places like Quantedia where you can find pre-made and back tested strategies on multiple different markets traded.
This is an absolutely insane resource for anybody looking to get started and there are multiple websites just like this. Next, I'll use Google Scholar to go away and research trading ideas. This is also a mind of information to try and find papers that have been well researched about trading and trading strategies. Also, another one is Trading View. Although there is a lot of garbage on Trading View, there are quite a few channels out there that actually do have a lot of value.
One of them being the NNFX channel where I have found most of my profit. Now whilst researching over the past 5 years I have found that two types of strategies have been the most profitable for me. First being mean reversion but the most profitable has been trend following. NFX go through a full system and share all of their indicators that they use on this website. Now this has been one of the most valuable websites that I found.
Building strategies using the NNFX method has been absolutely amazing. Now, I don't always use them for the actual indicators because Trading View is where I find a lot of my source. Trading View has probably the biggest library of open-source indicators which you can just simply import into your trading strategies and back test them thanks to tradingkit.com and your AI. So, as I said, I've been trading the NNFX system with AI for about 5 years, and it really has been the most profitable trading setup.
I built something called a boiler plate, which you can find more information about in my school down in the description, but basically what it does is it takes one indicator for our trend direction, two indicators for confirmation. Now, we're maybe looking for something like a MACD or an RSI. Personally, I don't trade those ones myself, but they are there just for examples. Then we add a volume or volatility indicator to help us stay out of the market when markets are going sideways.
And our final indicator, which I have my own favorite, is a stop-loss and take-profit indicator. I personally use the ATR bands as they tend to adapt to market volatility and are way better in back tests than things like a fixed percentage. So that was my secret source for my trading strategies. I built a full framework around that and it is very very simple for you now to take what I've just said, take it over to AI and ask for it to build the same one.
Now one thing that is important when building for a smaller account is your position sizing. When you only have 1,000 or less to trade with, one of the most important things is preserving your capital. In my boiler plate, one of the most important indicators is around position sizing. Making sure that my positions are much smaller when markets are moving sideways and much bigger when we're in explosive trends. So, with all that being said, how do we go now and take all of those rules, take them into Claude and ask it to start finding profitable trading bots.
Okay, for this to happen, I have created this extensive prompt here that it goes through in an effect and creates what we call a loop. What loop does is it loops through an idea multiple times throughout a scheduled period, building new strategies and optimizing them to find the most profitable setups or this continuous strategy research optimization and back testing loop. I have given it 24 hours. Now you can change that 24 hours to a couple of hours, 8 hours, 10 hours or a week.
It is up to you. It is going to go away and find strategies continuously and putting them into a mission control. That mission control will be hosted locally as a HTML file. So you can go back and see the equity curves and copy the code over to Trading View. So I'm going to hit enter on this. This is going to go away and actually start back testing strategies. Once it's done, you should have a similar control panel that looks like this where we can see equity curves.
You can see what it's actually trading, what co what time frame it was optimized on, things like win rate, profit factors, and all of that good stuff in one single place. Now I've made mine a little bit more sophisticated as I can go all the way through and see the live data and incubation which I'll talk a little bit later and you can see all of the information that's very very important as an algo trader but first of all you'll get something very very simple and just by talking to chat GPT or claude you can easily make these changes and create your own mission control like myself and the students in my school.
Okay. So once you've actually back tested those strategies and you have something that looks profitable, believe me, you will actually find something. AI has absolutely revolutionized this industry. Before it used to take me multiple days to actually find a profitable strategy. Now, thanks to AI, I'm finding new strategies every single day. Our next step is one of the most important. The next step is called incubation.
The problem with a back test and optimizing a strategy is that 95% of those strategies when put into live trading will absolutely fail. Now, I can tell you this from the bottom of my heart. I have back tested thousands and thousands of strategies and currently I'm only trading seven. Now, on my website, there are over 400 strategies which are fully incubated and still profitable. But like I said, I have back tested probably several hundred,000 settings.
So, incubation is leaving the strategy time to breathe. We're going to put it into our trading journal, part of the mission control, and leave it to see data that it's never seen before. So, that period could be 1, two, maybe 3 months, just sitting on your computer without actually doing anything. We then go back to Claude or Chat GPT and ask them to reback test all of those strategies. Now, personally, I do this daily, but you can just do it every week, every month.
It's up to you. When you ask Claude to reback test those strategies, ask it to separate the data into incubation and back testing. And you should end up with an equity curve that looks like this. This one here for example, green is back testing, pink is live trading or the incubation period and you should be able to move between backwards and forwards between the back testing and live trading data. This gives your strategy data that it hasn't seen before and helps you to judge whether or not it will actually work in live trading.
So once you've done that, you're going to go through something that I wished I knew as a trader five years ago and would have saved me tons and tons of money. I know it's hard to sit through 3 months and everybody just wants to get to trading, but honestly, if you only take one thing from this video, incubation is absolutely essential to becoming a profitable AI and algo trader. Right. Okay. Now let's say that you have a profitable bot that has been running for over 3 months on live data back tested as far as it possibly can.
The equity curve looks great. The draw down it doesn't look like it will explode your account. What are the next steps? This is where trading view becomes essential and I feel really good value for money by delegating the signals over to trading view. You don't have to have a computer that's running 24/7 and you can run on their infrastructure. So you simply go over to Claude and ask it to give you the Pine script code.
You go over to Trading View, click on the Pine editor just here and this will open this panel on the right hand side. We're going to paste our code in like I have just here and that will add the Trading View strategy onto your Trading View. You can see you can back test the strategy using the trading view back tester. You can change the dates, reback test. You get all of that amazing stuff that you would expect from a great charting platform.
But not only that, Trading View has something called an alerts system. Now, this alert system allows you to ping your broker's server and actually trade live money. So, this is where I'm going to go over to my broker. We're going to need to be able to connect our broker over to our Trading View account. Now, this is pretty simple for most brokers and pretty much is the same process for each and every one of them. First, we're going to need to go away and get an API key and secret from our broker.
Now, this is a bit like having a username and a password for your AI agent and doesn't give access to your full account, but simply allows it to trade. So, I'm already on my broker just here. Generally, you go over to your avatar just here and click on API. We then create a brand new API key and secret by clicking here and clicking here. Now, it depends on the broker that you're using. Generally, we just have to fill out a few questions.
Ask for read, write, or read trade permissions. We then go down and check a few boxes. Now, as an AI trader, we're going to need to be using unified accounts or futures accounts. That all depends on what you're actually trading. We're going to go down. We don't need access directly to transfer funds out. We're going to click on submit. That will ask for our 2FA validation. I won't be entering mine today, but I would show you the full process.
This will now pop up with an API key and secret which will go over to our middleman service. So, as I said earlier, the middleman service is a bit like a translator. It sits between Trading View and your broker and just literally translates the signal sent from Trading View over to the code that was made for your broker. I'm going to click here and sign in. I'm going to choose my broker from the list. Enter the API key and secret that I've received from my broker.
Press continue. And that will generate this code here. Now, all we have to do is copy that code over to Trading View where we're going to be setting up an alert just here. We click on the three dots where it says more. Click on add alert for the name of your strategy. And we're going to paste it into the message section just here. Now, when I said code earlier, this is the code that we are going to be sending. This will send directly over to your exchange and translate everything for you.
It allows for your exchange to have the entry, stop-loss, and take profit all in one single message so that you don't have to worry about your so you don't have to worry about your trades being set up and sent. Don't have to worry about stop- losses and take profits and things like that. We're going to give it a name. We're going to press apply. We will also need to enter a URL just here which is given to us by trigger trade just here.
We're going to paste that as well. Press apply. Press create. And Bob's your uncle. If you click here, you'll be able to see that we have a new alert. And that's it, guys. You can now sit back and watch your AI agent or your AI trading strategy trade for you. The only thing that you will need to keep an eye on is actually whether or not your strategy still has an edge because strategies are not made forever and eventually they will lose their edge.
Some strategies will last a couple of months and others will last years upon years like some of the ones that I have on my website. Now, even better, personally, I use this MCP server here which actually downloads the results from my account and Claude actually can make the decision for me. I leave the bots to run for a couple of weeks and then I ask Claude to keep an eye on those strategies daily. And thanks to this MCP server, it can actually turn on and off your alerts inside of Trading View, which means that AI manages your full portfolio for you without you ever having to be there. from building, back testing, creating those strategies to actually managing your account as a full handsoff approach, sitting back and just trusting the process, which obviously is where I'm going to segue over to my five rules.
Now, I've been back testing and creating and live trading strategies for over 5 years here on YouTube and have become one of the top traders multiple times on multiple brokers. Now, here are a couple of the rules that I wish that I would have learned earlier. one that I spoke about earlier which is incubation which is very very important to understand and really implement in your algo trading career. Next is the law of large numbers.
When you launch a bot, if you are putting it onto your live account, make sure you give it enough time to actually prove itself. Don't skip from one bot to another, otherwise you will end up just burning cash. Number three would be always back test your strategies because just trusting any AI agent or bot is something that I did in the in the past and that cost me more than anything else. Number four, although I said earlier about law of large numbers, it is okay to turn off a bot.
Make sure you have a full system in place to understand when it is time to turn off a bot. I personally use things like standard deviation around my equity curve and last year I personally ignored all of my rules and that cost me money too. Don't get emotionally attached to a bot no matter how profitable it was in the past. If all the signals are telling you to turn it off, it is better to turn a bot off now and then turn it back on later if it starts being profitable than it is for the equity curve to do a full home run on your profit.
And finally, no matter what people tell you, if you can find people to help you build strategies as a community, I found that that is way better than any AI system out there. Bouncing ideas off of each other is essential and helps you to think about things different. Personally, I've been invited to roundts where we've talked about things like position sizing, which has absolutely changed my way of thinking of trading and helped me to become much more profitable.
Now, if you don't know anybody else, you can find our link to our community down in the description down below. Now, of course, you can do everything by yourself from A to Zed by following this video and for free. But if you're looking for like-minded people that are building their future in AI trading, you're more than welcome to come and join us down below. And guys, without further ado, keep safe. Trade like a boss. >> Come on, David.
And once again, he's gone to a new strategy. He'll be back in just 5 minutes. RSI and MACD. If it doesn't run into any bugs, there will be a new strategy.
The words are the caption track's own and nothing is reworded or re-transcribed. Paragraph breaks are placed between sentences so the text reads as prose.
Free tools for your own script: paste a draft and see where it stands before you record it.
Paste your draft and see where viewers are likely to drop off, with a rewrite for each weak line.
Paste the first 30 seconds of your own draft for a hook score and rewrites.
Check your draft against YouTube's advertiser-friendly guidelines before you record it.
Read this channel's public videos and transcripts, and download a writing brief for it.