Used to detect fraud, run customer service chatbots, assess loan applications and even make investing decisions, AI has begun to do a lot of heavy lifting when it comes to fintech. It’s even helped financial regulators find the information they need in the U.S. Code of Federal Regulations; For context, the equivalent of parsing through the 600,000-plus words of chapter one is like reading “Moby Dick” three times in a row.
That’s why when Jay Artz, a senior principal architect at Vestmark, a software and services platform for wealth management firms, needed to speed up the data visualization process for customers, he turned to AI.
“Without Claude, this project would likely still be on the long list of great ideas we don’t have time to develop,” Artz said.
Vestmark provides wealth portfolio management and trading solutions for financial institutions and their advisors.
What types of products or services does your engineering team create? What problem are you solving for customers?
At Vestmark, we provide a wealth management platform that enables our customers to efficiently manage and grow investor portfolios. Our solution helps our clients streamline operations and focus on delivering value to their investors.
My team is responsible for the ingestion and analysis of financial records from partners each day, ensuring the data is accurate, reconciled and ready to trade. There are many moving parts to be orchestrated, executed and monitored to maintain the highest data quality. A single client environment can have hundreds of tasks running daily to keep the system operating smoothly.
By automating and monitoring these complex processes, we help our clients focus on decision-making and investor outcomes instead of operational overhead. This transforms a previously time-consuming back-office challenge into a reliable, scalable and transparent workflow.
Tell us about a recent project where your team used AI as a tool. What was it meant to accomplish? How did you use AI to assist?
With hundreds of daily tasks for each client, it was difficult to visualize progress or identify areas for improvement. Using the AI tool Cursor, I built a script that automatically generates a timeline image of each process. Previously, a timeline could be created manually in Excel — a time-consuming effort. With the script, a simple query now produces the image, saving hours and making the data easy for anyone to understand. Cursor made development fast; what could have taken weeks was completed in an afternoon.
The script worked well but still required manual queries and lacked interactivity. Later, using prompts with Claude, I created an enhanced web app in about 16 hours. The web app live-queries data and lets users drill down from a summary chart all the way to individual tasks. It provides clarity and insights that were previously unavailable, turning a complex process into a transparent, data-driven view.
“What could have taken weeks was completed in an afternoon.”
What would that project have looked like if you didn’t have AI as a tool to use?
When I asked Claude to estimate the development timeline for the web app we had built, it produced a six-month project plan for a full team. We built it in a week.
Without Claude, this project would likely still be on the long list of great ideas we don’t have time to develop. With AI, we delivered immediate value and turned an idea into a working prototype in days.
Debugging is another area where Claude shines. Normally, a single issue can consume hours of trial and error. Claude tirelessly tests different approaches until it finds a solution, making it faster and easier to work through problems.
Looking ahead, I see that AI will enhance our ability to add new features to our product. These tools can feel almost magical in how quickly they deliver results, but they don’t replace solid design and architecture. Success still depends on clear requirements, sound data models and scalable designs. AI simply shortens the path from concept to reality.
