In the world of artificial intelligence at TD, Sandra Aziz is a true pioneer.
Last year, Aziz, alongside a team of scientists, engineers, and data pros at TD, was heads down developing what they believe could be one of the most powerful AI banking solutions created in Canada.
Aziz — Senior Machine Learning Engineer at Layer 6 — and her colleagues created the Bank’s first agentic AI application, a new frontier of AI technology that is so powerful it can complete mundane tasks and solve complex problems with minimal human intervention.
In time, Aziz believes that agentic AI will help to revolutionize the banking experience for TD clients, potentially making everything from opening a new chequing account to applying for a line of credit easier and faster.
Agentic AI solutions at TD
“We built where nothing else existed. Everything is new,” Aziz said. “The agentic AI agent will summarize different information from different data points and provide the right information to a human.”
TD clients are benefitting directly from some of the agentic AI advances being developed by Aziz and her colleagues at the Bank.
In May, TD announced the launch of its first agentic AI model to help automate the application process for mortgages and Home Equity Lines of Credit.
When clients apply for a TD mortgage or TD Home Equity FlexLine, TD is using agentic AI to scan client documents, calculate their income, and validate it against policy requirements, perform consent checks, verify client income, and search for any discrepancies in the application before it automatically generates a memo for underwriters with its findings.
The results TD has seen so far suggest that it can take an application process that once required an average of 15 hours per client and complete it in an average of less than three minutes.
Aziz believes that Bank decisions and processes that used to take days will soon be able to be made at lightning speed thanks to agentic AI applications.
Aziz and her team at Layer 6 — the Bank’s AI centre of excellence responsible for developing cutting-edge technology solutions for TD — started by integrating agentic AI into the mortgage application process, but will soon be introducing the technology to help colleagues better serves clients end-to-end in every step of the mortgage process, from when a customer submits their documents to when funding is paid out.
These solutions, like all AI solutions at TD, are developed with the Bank's Trustworthy AI approach — a governance framework that includes checks and balances every step of the way and a human-in-the-loop.
Aziz and her team are actively developing agentic AI, using many of the foundational capabilities from this first solution throughout other areas of the Bank.
Growing alongside this new technology
Aziz has been on the frontlines of technological change at TD since she joined the Bank in 2014 as an intern armed with a master’s degree in statistics from the University of Montreal.
“When I joined, we were called the statistical team, and then we became more of an analytics team doing predictive modelling, and then we started to embrace AI,” Aziz said.
“I was introduced to more programming languages and to more complexity in the data. We went from basic statistical models to predictive modelling and then from generative AI to agentic AI.”
Over the course of her career, Aziz has been involved with several key innovations at the Bank, including her most recent project at TD Insurance where Aziz was leading the fraud analytics team.
“We were trying to capture a specific scenario about how to identify insurance fraud,” Aziz said. “Eventually, we ended up deploying a predictive model that can identify insurance fraud before it occurs, which was huge.”
This was the first instance that Layer 6 and TD Insurance collaborated on a project. That collaboration eventually led to Layer 6 and TDI introducing a client-facing chatbot, the TDI Virtual Assistant, earlier this month.
Aziz’s experience working with Layer 6 on that large and successful project eventually resulted in her joining the Layer 6 team in March 2025 as the Technical Product Owner, leading development of the Bank’s first agentic AI solution, one that speeds up the mortgage application process.
The cross-functional team Aziz led is comprised of research and data scientists as well as specialists from across TD who have a deep appreciation of existing Bank processes and databases, which are being used to help model the new AI applications being built.
AI technology will continue to evolve to make once lengthy processes simpler and faster. But one thing that won’t change is the Bank’s human-first approach, both when it comes to AI oversight and client-centred development.
“We have been able to make things move faster. But we also want to be able to make sure that whatever we’re doing makes sense from a client point of view,” Aziz said.
“We want to be first-to-market with the fastest processes and the best technology. But we can never lose sight of our clients, and what they need to make their lives easier.”
Want to learn more about Innovation?