Borderless AI is generating millions in revenue with an engineering team far smaller than what would traditionally be required to build and operate a global employment platform. CTO Derrick Isaacson says AI is a big reason why.
As Chief Technology Officer at Borderless AI, Isaacson leads the technology and engineering organization behind a platform spanning AI, global payroll, payments, and employment compliance, supporting companies including Mejuri, The Ritz-Carlton, and Ground News.
The approach reflects a broader shift underway in software development. Rather than simply adding AI features to its products, Borderless AI uses AI throughout the engineering process itself, from writing and reviewing code to testing, debugging, documentation, and product development. The result, Isaacson argues, is that a relatively small group of engineers can now produce the kind of output that once required a much larger organization.
That philosophy is informed by Isaacson’s career building technology teams designed to operate at scale. Before Borderless AI, he served as Director of Engineering at Instacart, where he helped establish the company’s Toronto engineering presence and worked on its white-label e-commerce technology. Earlier, he held engineering leadership roles at Neighbor and Lucid Software, as well as technical roles at Microsoft and Amazon. He studied distributed systems at Stanford and holds two technology patents.
Now based in Toronto, Isaacson is applying those lessons to a different kind of scaling problem: how to grow a technology company without automatically growing its engineering headcount alongside it.
The question has become increasingly relevant as AI coding tools reshape the economics of software development. For Borderless AI, the opportunity is not simply to make individual engineers faster, but to rethink how engineering teams are structured, what work should still be done manually, and how much leverage a small group of highly capable people can create.
In this conversation, Isaacson discusses how Borderless AI uses AI across the software development lifecycle, how a small engineering team can support a business generating millions in revenue, why Toronto has become an important technology hub, and the skills he believes will separate great engineers from everyone else as AI becomes a standard part of the job.
Why build your engineering org in Toronto, and what does it take to scale a world-class team here?
DI: Toronto has quietly become one of the premier tech and AI hubs in North America. Having established an engineering org here at Instacart and scaled engineering teams across companies like Lucidchart, Neighbor, Amazon, and Microsoft, I knew Toronto was the ideal location for our tech headquarters. We are building high-availability distributed systems operating globally, which requires top-tier talent in fintech, systems architecture, and applied AI. Toronto gives us direct access to exceptional engineering talent that knows how to build enterprise-grade, bulletproof software without sacrificing velocity.
As a CTO building in the post-ChatGPT era, how is Borderless AI actually leveraging AI within your platform and engineering org?
DI: We view AI as a foundational architectural layer, not just a bolt-on feature. Being built in the post-ChatGPT era gave us a clean slate to embed agentic workflows directly into our global fintech, payroll, and compliance engines. On the product side, AI powers our synthetic API integrations with global payroll providers and customer guidance through global employment relationships. On the engineering side, we use AI-driven tools to accelerate development velocity while maintaining strict engineering standards. When you are moving real money for global workers, AI provides the intelligence, but robust distributed systems guarantee deterministic reliability.
Global payroll and payments are notoriously complex. What engineered solutions make your real-time payment engine unique under the hood?
DI: It comes down to purpose-built distributed systems architecture. We engineered a proprietary real-time payment (RTP) engine that moves millions a week. To achieve that, we solved core technical challenges that legacy platforms struggle with:
- Dual-State Machine Technology: We developed proprietary state machines that automatically detect, retry, and recover from payment failure states without human intervention.
- Novel Priority Queues: We built custom-engineered queue management to create an instant payment experience even in corridors where local banking rails don’t natively support RTP.
- Unified Payment Pipeline: Rather than stitching together separate platforms, we engineered a single payments engine that unifies invoices, payments, and other transactions for both contractors and EOR employees globally.”
What is your philosophy on balancing rapid engineering execution with the strict reliability needed for global financial infrastructure?
DI: Velocity and rigor aren’t mutually exclusive; great engineering architecture enables both. Having grown engineering teams from 11 to 60+ engineers and scaled enterprise solutions for partners like Costco, I’ve learned that speed comes from getting your core abstractions right early. We integrated AML screening, account validation, and regulatory logic directly into the platform rather than treating them as external checklists. When your underlying infrastructure is clean, scalable, and self-healing, your engineering team can ship fast without taking on technical debt or putting customer payments at risk.
AI moves ridiculously fast. How do you ensure your engineering architecture maintains a technical moat rather than getting obsolete in a couple of years?
DI: In fintech and AI, if you stand still, your competitive advantage evaporates. Having designed distributed systems at Amazon and Microsoft and scaled engineering orgs through hypergrowth, I know you don’t build a moat by using the same solution for years. You build it through a continuous innovation loop and smart frequent investments in new techniques.
Our real-time payment engine gives us an immediate edge, but we maintain that advantage by automating each manual action we identify as the business scales. We engineered our system architecture in Toronto to be highly modular and extensible. That way, as AI capabilities advance, we can continuously upgrade our intelligent workflows without disrupting our underlying bank-to-bank payment rails.
Q: You’ve built systems at tech giants like Amazon and Microsoft, as well as high-growth startups like Instacart and Lucidchart. How does that experience shape the engineering culture you’re establishing in Toronto?
Derrick Isaacson: “Big tech teaches you how to build fault-tolerant, high-availability distributed systems at massive volume, while startups teach you speed, customer empathy, and ownership. At Borderless AI, we bring those two philosophies together.
Because we are moving real money for workers across hundreds of countries, our Toronto engineering team operates with a big-tech rigor around security, real-time risk decisioning, and system resilience. But because we built natively in the post-ChatGPT era, we operate with extreme startup agility. We give our engineers direct ownership over complex global fintech problems, combining world-class distributed systems engineering with applied AI to ship enterprise-grade platform capabilities in weeks, compared to quarters”