Given the recent trend of digitisation, using machines to invest doesn't come as a surprise, does it? But it certainly did surprise me! Considering the market is scaling fresh peaks and the possibility of a bubble bursting, machine-based investing could prove to be extremely beneficial.An extreme board game enthusiast with over a decade’s experience in finance and investing, Kanika Agarrwal, combined her interest with ambition and co-founded Upside AI - a portfolio management service company that uses machine learning to make fundamental investing decisions.“Upside AI really started as a science project,” Agarrwal recalled, walking down the memory lane.The trio - Atanuu Agarrwal, Kanika Agarrwal, and Nikhil Hooda - started Upside AI in December 2017, combining their interests in technology, finance and investing with their individual entrepreneurial ambitions.“Atanuu and I have spent our careers in investing, and Nikhil was just finishing up his PhD in machine learning. In the middle of a board game session, we came up with the idea of building something that learns how ‘fundamentally good’ is defined in the market,” she said.In reality, how machine-based investing works is a real quandary to some.The very foundation of Upside AI is the belief that technology will make better decisions than humans in long term since machines are unbiased and unemotional decision-makers.Also Read | How to unlock real-world business value from machine learningUpside AI uses technology to understand, recognise and buy companies that are fundamentally good businesses and also in-demand stocks, said Agarwal who loves rewatching the Lord Of The Rings trilogy every year.“For us, we are using machine learning to read company fundamentals,” Agarrwal said.The question an investor would now ask is — how is machine-based investing any different from the manual form of investing where fund managers put their decade-long experience to use.“The main benefit of our model is that it is dynamic. It is trying to understand the meaning of good fundamentals in the context of the market,” Agarrwal explained.“This means, it is not restricted in its approach — it can be a growth investor, value investor or a commodities expert depending on where the winds are blowing,” she added.Also Read | Adoption of game-changer technologies by payments industryFurther, the lack of bias or emotion means the machine has no baggage from its past success and failures, unlike money managers, remarked Agarrwal who has worked with companies like Mayfield India, Credit Suisse and EY.On being asked whether machine learning would mean the reversal of a passive investment trend, she firmly denied it.In fact, Agarrwal went on to add, “Over the next decade, just like the US, rules-based products will become mainstream. This includes passives from asset management companies and active products like ours”.Lastly, when asked Agarrwal about her investment strategy specifically when the market is hovering near record-high levels, the voracious reader said there are pockets of over and undervaluation.Agarrwal added that after Q1 earnings, the Indian benchmark indices are trading at very reasonable valuation multiples with significant tailwinds.“Having said that, we don’t take short-term calls on the market,” she replied.“Long term markets trend up, not down and we are in it for the long run. Case in point, we are fully deployed today because we do not believe in timing the market,” said Agarrwal who aims to build Upside AI into India’s first fully tech-enabled asset management company.Also Read | Does Day trading as a side hustle make money?Upside AI ranks among the top-performing portfolio management services in India delivering 71 percent cumulative returns since July 2019. In June 2021, PMS AIF World ranked Upside AI Multicap in the top 10 products of its peer group.On the back of performance and organic referrals, Upside AI's assets under management have grown by 10 times over the last year to more than Rs 90 crore with funds from several high net worth individuals and family offices.