When Shary Mudassir joined RBC (Toronto Stock Exchange: RY-CA) Capital Markets eight years ago, seven out of 10 junior hires in the equity trading unit had a business background. Now, 70 percent of newbies are engineers.
Speaking on Wednesday at an event in Silicon Valley, Mudassir said his group is staffing up on programmers to help Canada's biggest bank move from the era of electronic trading to the age of machine learning. Mudassir said he has a team of five people focused on building machine learning models.
Machine learning is a set of techniques by which computer programs can improve the answers they give over time without requiring programmers to change the underlying code -- instead, programmers "train" the programs by feeding them massive amounts of data, and they improve on their own.
So how does this fit into banking?
It's no longer just about getting clients the fastest trades at the best prices. Customers want much more information, including data on counterparties to help judge the risk of doing a deal. For example, is a big buy order coming from a giant mutual fund company or a high-frequency trading shop?
RBC can combine its massive data sets with custom code to help provide answers to these sorts of questions.
Mudassir, RBC's director of global algorithmic trading, said that competition is fierce from hedge funds and the other big investment banks as well as emerging technology companies that have their eyes set squarely on Wall Street. Mudassir didn't provide names, but venture-backed companies applying machine learning to trading include Sentient Technologies, Alpaca and Walnut Algorithms.
"Unless we can roll our sleeves up and really improve our game, it's probably open competition," Mudassir said. "Someone can come in, rightfully so, and provide a better solution. But we do have that domain knowledge and hopefully we don't get lazy."
Mudassir joined other RBC executives on a panel in Menlo Park, home to many of the world's biggest venture capital firms. The topic was machine learning in capital markets, and the panelists discussed their expanded use of technology and machine learning algorithms across the bank, from sales and research to compliance and operations.
It's a hot-button issue in finance. On Tuesday, BlackRock (NYSE: BLK), the world's largest money manager, announced plans to shake up its business of actively managed funds, cutting jobs and fees and turning more to computers for stock picking.
According to a report last year from Goldman Sachs, machine learning and artificial intelligence will enable $34 billion to $43 billion in annual savings and revenue opportunities within the financial sector by 2025.
Mudassir said his group has almost a dozen open positions that are difficult to fill because of the lack of talented programmers who also have a passion for finance.
"You have to be able to enjoy what you do and have that technical expertise," he said. "It's very fast-paced." Traders may no longer have to come in at "the crack of dawn, but it's still pretty early," he said.
More From CNBC