DURHAM – Through the use of artificial intelligence, Durham-based Trill Financial is changing the way asset managers gain a competitive edge by using artificial intelligence.
Trill Financial was founded in 2014 by Akash Ganapathi and Simon Jung, who act as chief executive officer and chief technology officer, respectively.
After graduating from UNC-Chapel Hill, Ganapathi worked at Cisco Inc. and SAS Institute Inc. as a software engineer and also worked with predictive analytics. Jung also graduated from UNC-CH and later worked at Cannon Research Center as a data scientist.
During their sophomore year and through their senior year, Ganapathi and Jung traded foreign currencies as a hobby, using algorithmic models to make almost 50 percent returns during one six-month period, Ganapathi said.
But, the two only had a few thousand dollars to their name, lacking the resources to scale their accomplishments trading foreign currencies into a large-scale business.
Once the two graduated and took jobs, and after taking some online courses on machine learning and deep learning from Stanford University, the Georgia Institute of Technology and the University of Michigan, they decided they had the knowledge to start their own company.
“It was a long road, even from there, where exactly we can find our best fit,” Ganapathi said.
“We found it now.”
Currently, there are hedge funds using artificial intelligence to identify market patterns on a weekly or monthly basis, such as the price action and volume of trading, Ganapathi said.
According to Ganapathi, they’re using data in a quantitative way, so it doesn’t translate well into the long term.
Trill Financial looks to make itself viable in the long term by aggregating data from the Securities and Exchange Commission, such as income statements, balance sheets, cash flow statements, earnings statements and other relevant data that allows users to create models for all publicly traded U.S. companies.
The way these models are created is a form of AI known as deep learning. Trill uses deep learning, essentially, by feeding data into a software that will then predict future markets.
The technology is currently being utilized by companies such as Netflix, Google and Amazon.
“We’ve been translating everything that’s been coming out in the deep learning space over to the finance space,” Ganapathi said.
Deep learning is the latest facet of AI, allowing portfolio managers to use the data they look at everyday and analyzing it in a predictive way.
“There’s almost no firms in the long term space that are really experimenting with AI in a meaningful way. It’s really in its infancy and we’re poised to really accelerate the adoption of AI in the long term space,” Ganapathi said.
The software that Trill offers is also customizable, allowing asset managers to create their own models easily to hopefully drive lower costs and better overall performance.
Once the models have been made, Trill is then there to provide support to customers to make sure they understand the product and its use.
Since its founding in 2014, the company has completed four rounds of private equity offerings.
On Oct. 19, 2017, Trill completed its fourth round, raising $370,000. The three previous rounds raised $350,000.
Trill’s investors up to date have been angel investors, but it plans to seek its first round of institutional investors within the next one or two years.
Trill plans to use these funds primarily to increase customer traction. Ganapathi said the company has a number of pilot customers and they hope to expand on that, while also converting pilots to full-paying customers. Expanding its customer base also means insuring it has the bandwidth to serve a larger number of clients.
According to Ganapathi, a few of Trill’s pilot companies are top 10 U.S. asset managers. Customers also include smaller hedge funds as well.
Ganapathi said the interest large asset managers have shown that Trill’s product is viable as a long-term solution. But, since they are one of the first to utilize AI in this way, it’s still a learning process.
“We’re finding out exactly how we can fit into their current process and help them down the road as well,” he said.
“We’re learning with them, they’re learning with us.”
Though Trill is still in its early stages, Ganapathi sees the company as the instrument that allows asset managers to have portfolios run almost completely by AI.
The two co-founders bring their knowledge of AI to the table, while a team of financial industry experts such as Eric Freedman, chief investment officer at US Bank; Thomas Looney, former Lenovo vice president; and Kevin Adams, a senior vice president at Raymond James Financial; sit on the board of directors.
Ganapathi and Jung connected with the current board members through their network of investors and friends in the Triangle area. Board members have given feedback on the business model and product, and also the overall direction of the company, while opening doors and providing introductions to potential investors and customers.
“They took a very early interest in a very early state company,” Ganapathi said.
Ganapathi credits their early interest to the potential Trill has to change the financial industry.
Ganapathi believes Trill Financial will change the financial asset management industry, and the industry looks susceptible to change.
According to the YouGov Affluent Perspective 2017 Global Study, “desire for immediate action has motivated one in five affluent consumers to turn to automated investment services, or ‘robo investing.’”
Affluent millennials show an even greater tendency towards automated investing, as 36 percent use digital tools to conduct their investments.
The study also showed that 61 percent felt computer-backed investing was the future, and 41 percent saw no reason for an investment advisor.
Though Trill is currently working with asset managers, Ganapathi hopes to eventually expand its platform to investment products that more people use, such as mutual funds and retirement funds.
Note: This story is from the North Carolina Business News Wire, a service of the UNC-Chapel Hill School of Media and Journalism.
VISIT THE SOURCE ARTICLE
Author: || World Economic Forum