Before setting up quantitative robo-adviser qplum, Mansi Singhal was a trader at Wachovia, Bank of America Merrill Lynch and Brevan Howard. She said the idea came about as she became increasingly frustrated with the mismanagement of her own personal finances, which prompted her to leave her trading position to become a fintech entrepreneur.
While the large majority of robo-advisers either court individual investors to grow their assets under management or provide automated investment management software to wealth management firms or defined contribution retirement plans, qplum is targeting both institutional and retail investors. It is based on a machine-learning approach to building trading algorithms, making it a more like a quant-style hedge fund.
From academia to the sell side to the buy side to a startup
Singhal’s has both a background in computer science background combined with experience as a trader on the sell-side and the buy-side. While Singhal was finishing off her Master of Science in Engineering (M.S.E.) degree in computer science at the University of Pennsylvania, she interned at UBS in 2005.
From there, she worked as an options trader at Wachovia, specializing in OTC interest-rate derivatives, where she rebuilt the quantitative models for option pricing and hedging and automated risk reporting.
“It was very challenging, but I loved it,” Singhal said. “For the first year, it felt like it was such a steep learning curve, and if you’re on a trading floor, there is typically huge learning curve for the first couple of years, but it was the best learning experience, from a quantitative perspective.
“People saw my computer science degrees and thought, ‘She can do models,’ and when I look at it in hindsight, having an engineering mindset can be a huge advantage,” she said. “More experienced traders have an edge, their intuition or the benefits of having been in the trenches, but for a new trader to prove yourself, you have to have some edge.
“My computer science background and engineering mindset were my edge – even for people who studied finance, the real world is so different, so you have to learn on the job, and that is especially true for trading roles.”
Wachovia was acquired by Wells Fargo soon after she left.
Singhal joined Bank of America as a short-end swaps trader just months before the height of the financial crisis in 2008 when BofA acquired Merrill Lynch. She modeled and launched a new OTC options product and claims to have exceeded revenue projections in both 2009 and 2010.
“Especially during the crisis, there was so much volatility, and I could see opportunities, but you’re also constrained by what kind of risk you can take and what kinds of products you can trade,” Singhal said. “Regardless of what the market is doing, every desk has a revenue target, and you have to keep working toward it every day, finding the appropriate models and doing things right, rather than throwing darts.
“The crisis went on for years, but you have to trade every day and make sure you get a little closer to your revenue target – some people miss that,” she said. “People think a star trader or rainmaker is going to make a lot of money, but it’s not so easy.”
At that point in her career, she was ready for a greater challenge: running her own firm. Along with co-founder Gaurav Chakravorty, Singhal was the co-founder and portfolio manager of Circulum Vite, which she describes as a multi-asset class quantitative trading and technology firm specializing in applying mathematical modeling to the fixed income and FX markets.
“I realized that managing money and gathering assets are two different games altogether,” Singhal said. “If you look at any hedge fund, you’d be amazed by the marketer-to-PM ratio – typically there are at least twice as many marketers as PMs, because asset-raising is very difficult.”
Eventually, Brevan Howard poached her for its New York office in January 2014 as a trader on a $100m global macro mandate. Singhal got a lot out of the experience, but said that the environment was hypercompetitive.
“I was completely flattered, because it’s one of the most respected names in the hedge fund industry, and it’s incredibly hard to get a PM position there, so I was extremely excited, even though it is more of a macro discretionary trading shop,” Singhal said. “It has incredibly smart people, but that means you’re also competing with smart people.
“The biggest change, at a bank on a trading desk, there’s more than one banker trading a single product, all working toward the same goal, with the same target revenue,” she said. “It’s competitive, but you’re all still working toward the same goal, whereas at a lot of buy-side firms, it’s very competitive among PMs, which can lead to its own complications.”
The desire to run her own firm never left her, and so Singhal departed Brevan Howard and co-founded quantitative portfolio management startup qplum along with Chakravorty. Prior to the launch of qplum, the latter led high-frequency futures and forex trading as a portfolio manager and partner at Tower Research Capital before starting his own quant hedge fund, DV Capital.
Having just debuted last year, it is still early days – qplum currently has approximately $5m under management and close to 40 clients. The minimum investment is $10k for retail investors, but Singhal and Chakravorty believe that their strategies are scalable to institutional proportions.
“Our job here is to use machine learning based trading strategies and implement then in such a way that the complexity is made transparent through demystification – we want to stay connected with investors but use high quality investment management techniques behind the scenes,” Singhal said. “Institutional investors always ask us, ‘What is the capacity – can it be scaled up?’
“We’re building an infrastructure that can grow in a way that’s scalable so that when we’re executing trades, we’re not losing money in the slippage,” she said.
Photo credit: William_Potter/GettyImages; photo of Singhal courtesy of qplum