Upside AI: Where artificial intelligence is the new stock market expert

This Mumbai based investment start-up is revolutionising the trading industry by using machine learning tools to curate stock portfolios.  


Trading and investing in the stock market is for the bravehearts. RIP Rakesh Jhunjhunwala.

However, new times bring forth new trends and the stock market is no exception in this regard. Young people nowadays have found a new passion—investing in stocks via digital channels. But trading on the bourses is a risky business, even if you think it’s a quick way to make some money. That is why experts and even disclaimers advice people to remain cautious while investing or trading in stocks, equities and mutual funds. Retail investors too always seek expert advice before investing in anything.

For a long time, consulting and brokerage firms have employed experts to provide advice and direction to early investors on where to put their money. And for advice on investments, stocks, and other investment portfolios as well. Their personal experience, built on both profits and losses, has always been the foundation of their mastery.

However, the advice may not work sometimes. Various factors like the current scenario at the stock markets, a consultant’s strategy and their emotional biases and affinities can lead to the failure of certain predictions. In order to avoid stock market failures, a three-year-old fintech start-up has started using artificial intelligence to generate accurate predictions.

Upside AI claims to have revolutionised the trading industry with its AI-based investment services. SME Futures spoke to Kanika Agarrwal, one of its three co-founders to gain a better understanding of the dynamics of AI-based trading patterns and curated portfolios.

What builds trust?  

An American economist once said, “The investor’s chief problem, and even his worst enemy is likely to be himself.”  

In a country like India where investment has always been considered a personal affair and where people often lack any knowledge about how to invest, this sentence fits like a glove. Now the question is, will machine learning and its algorithms be able to replace human experts ever? How will the investors of India accept this technology over human expertise?   

Well, sufficient data shows that over long periods, human beings, even experts tend to underperform below the existing benchmark. According to a recently released S&P Global report for India, it has been found that 80 per cent of mutual fund managers have underperformed below the benchmark over the last 5 years.   

“Even Warren Buffet has publicly said that after he is gone, he would prefer the Berkshire funds to be invested in ETFs which track an index instead of active management,” Agarrwal tells us.   

She further adds, “The consequences of human underperformance are visible in a mature market like the US. The top five hedge funds are all quant funds. Further, there is more retail money now in passive funds than in active human-run mutual funds in the US. We will see this trend play out in India as well.”   


Machines over humans 

The founders of Upside AI believe in the adage of the ‘Oracle of Omaha’ aka Warren Buffet who famously said once, “To invest successfully over a lifetime does not require a stratospheric IQ, unusual business insights or inside information. What’s needed is a sound intellectual framework for making decisions and the ability to keep emotions from corroding that framework.”  

And that is exactly how this investment platform does AI-based computations to curate a database of the stocks available in the market.  

Agarrwal elaborates that their AI tries and solves the two-sided problem of demand and supply. There is a plethora of fundamentally sound companies in the market. The system analyses the financials of all the listed companies and determines which of these fit the bill of being fundamentally sound. 

“On the other side is the demand for stocks. Even if a company is solid, there might be no demand for its stock in the market at a given point in time. Thus, the system also determines the prevalent market conditions and determines which portfolios are best situated at the intersection of perceived and actual value,” she further adds.   

Here’s how it works  

When asked about how their platform exactly works, the founders tell us that it’s based on the simple mechanism of input and output.  

The financial data of all the NSE stocks is fed into the algorithm, following which, the algorithm learns through millions of iterations about how to pick the companies that are not only fundamentally good businesswise but are also in demand for their stocks, according to Agarrwal. 

After that, the machine suggests a portfolio of stocks based on its analysis and the portfolios are made. Later, the Upside AI analysts perform manual checks on corporate governance to ensure that these companies are reporting their financials accurately.  Finally, the platform suggests a portfolio of 10-25 stocks to buy in order to start the investment and then it re-evaluates the chosen portfolio every quarter. 

A board game marathon engendered this start-up 

Initiated in 2017, Upside AI believes that technology will make better investing decisions than humans over the long term. This is because machines are unbiased, unemotional and unaffected by market euphoria and panic.  

The three co-founders, Kanika Agarrwal, Atanuu Agarrwal, and his childhood friend turned business partner Nikhil Hooda, came up with the idea of inception while playing the marathon board games, Agarrwal reminisces.   

Kanika Agarrwal manages the sales and Nikhil Hooda looks after the technology, while Atanuu Agarrwal manages the investment part. Kanika claims that their platform scans the entire stock market 50 million times to discover the best investment portfolios.   

A smooth journey  

Since its inception, Upside AI has launched three products for HNIs (High-net-worth Individuals- a term used for super-wealthy investors) and done research for different ticket sizes and geographies. Kanika claims that they have experienced a 5X growth in the assets that they have been managing in the last 18 months and they are already expanding their team.   

She tells us that initially, their Mumbai-based setup had only three members i.e., the three co-founders themselves. But gradually they have grown into a team of 12 working from three different cities.  

Talking about their performance and revenue model, Agarrwal tells us that their start-up has received an extremely positive response since its inception, and they have served nearly 200 customers so far. “We have two kinds of customers, the first kind being those who have multiple PMS (portfolio management services) investments and want to diversify away from human biases and use rules to invest. And the second kind is first-time investors who already believe that machines will beat humans over the long term. Our business model is based on a mix of fixed fees and profit shares,” she informs us. 

Kanika also tells us that initially their company was bootstrapped. “For our business, we needed to get a PMS license. And the capital required for the license is in crores. So, being middle-class kids, our initial investors were friends and family who believed in us and our vision. And I am proud to say that we were profitable from day one,” she asserts. 

They received a seed round of venture capital last year from Endiya Partners, Dr Vijay Kedia, and a few other angel investors. “The main reasons for raising this capital were to scale up faster and to hire the right team to build our products and expand our distribution,” she says. 

Will AI be the future of the investment arena?   

In today’s globalised world, artificial intelligence has seeped into every arena and is an integral part of every kind of technology and the stock market and fintech are no exception. Even though the target audience of fintech companies like Upside AI is niche today, Agarrwal feels that in the coming decade, there will be an influx of such tech-enabled stock portfolio curators in India. 

“When you think about our stock market 10 years from now, do you think there will be more shops like us or fewer? Clearly, the answer to that is that more and more investment firms will start to adopt technology and the practice of using rules to invest will go mainstream,” she says. 

Talking about their future, Agarrwal says, “Given the fact that we have an early mover advantage, we will have the longest time to build a track record, improve and launch new products and gain a better understanding of the market.”   

“Over the next decade, we hope to build an asset management company where the products and processes come with a built-in technology,” she adds.   

Crypto: Uncertainty encrypted!  

While most of the world has wholeheartedly accepted cryptocurrencies and is actively trading in them, there is still an uncertainty about them in India.   

“As an asset class and an emerging technology, we believe that crypto can be extremely powerful in the coming decades. However, as with any emerging tech, there is a lot of uncertainty and volatility associated with it, which should be taken into account before investing in it,” Kanika points out. 

“Therefore, we recommend to our investors that they should put a very small portion (less than five per cent) of their assets into crypto. They must also realise that there is a possibility of their investment witnessing deep drawdowns given the volatility in this arena,” she suggests.   

Agarrwal further says, “Investing in crypto in India has the added complexity of taxes and uncertain regulations. Further, since you must hold your coins and tokens with the Indian exchanges and cannot transfer them to a different wallet, there is the added risk of something happening at the exchange level to your investment.”  

“Considering all these factors, we should be extremely cautious about investing in crypto in India today,” she says before signing off.