Mercor Developer AMA Series — Bot Capital

At Mercor Finance we always strive to partner with the very best. Therefore we partnered with Bot Capital.

Bot Capital is an experienced software- and developers firm, which reached returns of nearly 2000%. They are a prominent entity in the algorithmic trading world. Having built up experience in the algorithmic trading world for decades allows them to create bots that outperform the market almost every time.

Bot Capital partnered up with Mercor Finance from the very beginning. Ever since the release of the Mercor Early Beta, they deployed amazingly strong algorithms on the platform. Last week, some of their algo’s reached an incredible profit of almost 60%!

One of Bot Capital’s strongest algo’s

At Mercor, we understand the word ‘algorithm’ can sound complex and repellent. Therefore, Mercor values transparency and education highly. We like to keep our community up to date and as informed as possible. That’s why we launched a new Mercor series: the Mercor Developer AMA series. Given their extensive expertise and after achieving incredible results, we figured Bot Capital would be a fitting candidate for our first-ever Mercor developer AMA series.

In this article, we will set out last week’s AMA, held in the Mercor community chat. The article will give you an insight into a prime developer on the Mercor platform. The AMA was a great success and gave exclusive insights into the benefits of algo trading. Below is a detailed summary of what was covered in the AMA. Enjoy reading!


‘The Mercor developer AMA series are all organized according to the same procedure. The first part is a one-on-one interview between Maxim, CEO of Mercor, and the interviewee, in this AMA, Bot Capital. During the second part, community members have the opportunity to ask their questions to the Mercor Developer.’

Mercor CEO:
Alright, let’s start with the first question!

Please give us a brief introduction about yourself (background, experience).’

Mark | Bot Capital:
‘I’m a builder. I’ve been in the tech industry for 20+ years and previously worked in engineering and leadership functions for big tech in the areas of payments, finance, machine learning, and optimization. Background in comp science and math primarily.’

Mercor CEO:
‘Impressive CV with a lot of experience in different fields! So let’s dig a little deeper into the algorithmic trading background, how and when did you get involved with algorithmic trading?’

Mark | Bot Capital:
‘I started getting into crypto soon after Bitcoin was born and briefly ran my own node almost a decade ago now (sadly I didn’t keep anything from that time). I quickly realized that to trade crypto effectively required full attention, a lot of sleep deprivation, and a systemic edge. Around that time I began experimenting with and developing trading systems. Since then, the systems and algorithms have evolved and improved significantly to a point that nearly all of my personal trading became automated and I don’t need to check my phone every few minutes :).’

Mercor CEO:
Great to hear that you have been involved with crypto from the absolute start! And I think many of us can relate to crypto trading causing sleep deprivation. So let’s learn a bit more about Bot Capital now, what is the story and mission of Bot Capital?’

Mark | Bot Capital:

‘Bot Capital started with a simple mission: outperform buy-and-hold returns of top-performing assets in all market conditions.

We started trading Bitcoin and Ethereum futures as it gives the ability to open both long/short positions to maximize profits. We then expanded into other large/mid-cap assets, and a year or so introduced spot strategies (which are more or less a long-only portion of futures trading).’

Mercor CEO:
Great, I think Mercor users definitely experienced that your algorithms perform in all market conditions so far!

Talking about your algorithms, what do your algorithms focus on? How do the algorithms of Bot Capital differ from other algorithms?’

Mark | Bot Capital:
The algorithms look at both short and long-term trends, volatility, momentum, correlations with other classes, and have a small number of proprietary indicators. These all are assigned various factors to generate buy/sell signals. We also integrate with 3rd party on-chain systems (CryptoQuant, Glassnode, etc), and process a stream of on-chain data. In recent years the on-chain data was helpful in predicting over half of the larger price drops and enabled the algorithms to prevent drawdown before technical signals kicked in.

Mercor CEO:
‘Impressive! Especially combining on-chain and off-chain data causing for less drawdown. So as we all can imagine these strategies are not made overnight, so maybe you could elaborate on this, how long does it take to develop a strategy? And how many people are involved in this stage of development?’

Mark | Bot Capital:
It’s a continuous process, and improvements are made over time. It took around a year to get a high-performing baseline crypto strategy that incorporated various signals (backtested on all available historical data for Bitcoin). We calibrate the strategy (using ML and some optimization algorithms) to each traded asset to account for differences in market cap, implied volatility, and other factors. Once the base strategy is in place, calibration to new assets is pretty quick, just requires the processing of historical data.’

Mercor CEO:
‘So updates are continuous and algorithms are continuously improving, great to know ;).

Could you give us some more insights into your backtesting results and performance history?’

Mark | Bot Capital:
We measure performance on the scale of years and don’t look at monthly/quarterly performance to avoid recency bias and market fluctuations. For instance, for 12 months prior to launching on Mercor, the ETH spot strategy returned 1,908%, and ETH perpetual futures strategy returned 12,621% with no leverage (in both cases assuming 0.05% trading fees, which is a bit less than DEX). In the same period, ETH buy-and-hold returned 765%. In general, our minimum threshold for launching a strategy is to outperform (in backtests) buy-and-hold returns of whatever asset we’re trading by a factor of 2x (so that trading makes sense from a tax standpoint), annually, in both bull and bear markets. There were months that the strategies had a drawdown of 20% or so, but on average the results have been very positive over the long term.’

12 months prior to launching on Mercor, the ETH spot strategy returned 1,908%, and ETH perpetual futures strategy returned 12,621%

Mercor CEO:
These results are incredible and I admire your very strict threshold of outperforming the underlying asset by a factor of two. As far as looking at algorithms over yearly periods this is definitely something that we all should be doing more, algorithms need time and in your case have outperformed the market and underlying assets with impressive figures.

So let’s talk about Mercor, what do you think of the Mercor platform so far, its functionalities, the team, and its ambitions?’

Mark | Bot Capital:
A decentralized platform for automated trading is the first of a kind, innovative, and has the potential to disrupt traditional trading. The platform supports trading of most common pairs that you’d want to trade and just works. I look forward to the Mercor dev team adding additional capabilities in the future such as derivative trading, but that’s just a bonus for the future. The team is world-class to build a platform like this and has been iterating very quickly since the first beta to make important changes. Also very impressed with the quality of the beta launch. Very impressive!’

Mercor CEO:
‘Great to hear, especially coming from you with your experience! We will definitely continue to improve the platform and I cannot wait to show you and the community our roadmap for Q4 2021 and 2022.

Alright so that brings an end to the first part, the chat is now opened up for community members to ask any questions, if there are still questions after these thorough explanations ;).’

Meikol, Mercor Community Member:
I have some questions regarding the input of the model and its parameters. For the ML models which act on sentiment, where do you get your data from and how much data do you input for the model to train? How does the model actually measure sentiment (number of keywords per timestep) and is there a clear distinction in positive/negative sentiment?’

Mark | Bot Capital:
[In reply to Meikol]

We don’t look at sentiment beyond fear & greed index. The ML is used primarily to find correlations with other asset classes (clustering, etc), traditional markets, and to balance out the parameters of various indicators.’

‘I have some questions regarding the input of the model and its parameters. For the ML models which act on sentiment, where do you get your data from and how much data do you input for the model to train? How does the model actually measure sentiment (number of keywords per timestep) and is there a clear distinction in positive/negative sentiment?’

Mark | Bot Capital:
[In reply to Meikol]

‘The models are slightly different for bull and bear markets, and the strategies switch between modes when Bitcoin and some other assets fall under 21W EMA/SMA.’

Tom | Mercor Finance:
‘Hi Mark, one of the most impressive results of an algorithm since the launch of the Mercor platform was for me the reaction of one of your algorithms upon the Litecoin event. Could you please elaborate on how the algorithm achieved this impressive result? For the people who do not know what I am talking about, please refer to the video below:’

The Litecoin event, displaying the value of Mercor

Mark | Bot Capital:
[In reply to Tom | Mercor Finance]

‘That was a lucky catch. In general pump/dump and spikes/drops like this are very common in crypto, and the algo’s are optimized to detect over-extension and react to it. For instance, when LTC pumped 20% in minutes it reached unsustainable/overbought levels, and the algo’s sold at the top.’

Mercor CEO:
Definitely an amazing moment.’

Idan, Mercor Community Member:
Hi Mark, for people who want to start building trading systems. What language do you think is best?’

Mark | Bot Capital:
[In reply to Idan]

It depends what you want to build. Most folks these days use Pine in TradingView to build trading strategies, and the language and capabilities of TV/Pine are very powerful, you can build pretty much any technical indicator. However, Pine/TV doesn’t support external data sources like on-chain systems, and for that, you’d have to build a custom solution outside of TV, and can use any language you’re comfortable with.’

Mercor CEO:
‘Alright everyone, I will now close this AMA. Thank you @botcapital for this very informative AMA and thank you to all community members for taking part! If you enjoy these types of AMA’s we will definitely continue to organize them!’

This was the first of many Mercor Developer AMA’s to come. The community got exclusive insights into the methods of our beloved bot creators and got to chat directly with the creators of these profit machines!

One of Mercor’s main goals is to make algorithmic trading accessible for everybody by building a bridge between talented developers and expert algo traders and everyday investors. This AMA series is a prime example of doing just so!

The Mercor team enjoyed the AMA to the fullest and loved to see the engagement between investors and developers. On to the next one!

Mercor has just released its updated roadmap and it’s ambitious!

Check it out at:

Stay connected

Telegram | Twitter | Mercor Beta Platform | Website | Facebook | Instagram | YouTube