At ComplyAdvantage, scale isn’t a buzzword; it’s the environment you step into from day one. Our systems process hundreds of millions of screenings daily and help financial institutions uncover risks that could otherwise go unnoticed. As an engineer, you quickly learn that the problems you’re solving matter, both technically and in the real world.

I’m Joana, a Senior Software Engineer at ComplyAdvantage. Over three years ago, I joined as a mid-level engineer after working mostly on smaller products. I wanted to deepen my technical skills, and I knew I needed an environment that would challenge me. What I didn’t expect was just how quickly I’d grow.

A Steep but Supportive Learning Curve

The transition into CA’s ecosystem was exhilarating. I moved from working with modest tech stacks to an architecture built on Kubernetes, Kafka, Elasticsearch, MongoDB, Grafana and an API-driven distributed platform spread across multiple regions. Many of these technologies were things I’d studied or experimented with, but had never used at true production scale.

I expected to feel overwhelmed, but the engineering culture here is incredibly collaborative. The team is filled with people who love complex problems, think deeply about design, and are always willing to help. That combination of ambitious systems and genuinely supportive colleagues made the learning curve not just manageable, but energising. It’s hard not to grow quickly when you’re surrounded by people who are both experts and great teammates.

Rethinking Assumptions: Lessons from Different Teams

My first role was on the Ongoing Risks team, which owned the monitoring system behind our legacy customer screening product. The system had a simple goal on paper: re-screen every customer daily to detect new risk information. In reality, that meant processing hundreds of million customers every 24 hours across several global regions.

As usage grew, so did the pressure on the system. The core challenge became finding ways to scale daily processing without compromising on performance or drastically inflating costs. One of the most impactful projects I worked on was improving how work was parallelised, which shaved two to five hours off the daily run in our busiest regions.

Another major improvement focused on eliminating unnecessary searches against some of our Elasticsearch clusters. Most daily screens produced identical results, meaning the system often repeated expensive searches that didn’t change. By rethinking how and when data should be re-queried, we were able to dramatically reduce the load on Elasticsearch. This not only improved performance, but also allowed us to shrink cluster sizes by as much as 50-60% in some regions (confluence link), an enormous cost saving with zero impact on output quality.

These projects marked the first time I’d worked on systems operating at this scale. They also showed me that big improvements often come from rethinking long-standing assumptions rather than just pushing more infrastructure at the problem.

Today, I’m part of the Monitor & Search team, and the scope of the work is even more exciting. We own both the search engine and the next-generation customer monitoring system, the heart of how clients detect risk with our new product.

This team is a blend of software engineers and Machine Learning engineers, which makes the environment uniquely dynamic. On any given week, we might be evaluating search precision, redesigning a distributed workflow or exploring how machine learning signals can enhance our matching engine. Improving both speed and accuracy is a constant balancing act, and it’s one of the reasons I love this work: it combines algorithmic thinking with real-world engineering constraints.

We’re also tackling one of the most significant architectural shifts in our product: low latency Monitor. The goal is for clients to be alerted about relevant sanction risks without having to wait for the daily re-screens. Instead of waiting up to 24h, our platform will notify clients in a much reduced time window from the moment a sanction list is updated. This is exactly the kind of problem that forces you to think deeply about design, data flow and the long-term impact of each decision.

Looking Forward

What excites me most about working at ComplyAdvantage is that the problems never get smaller, only more meaningful. We’re building distributed systems that operate at global scale, improving search accuracy with machine learning, optimising costs in smart ways and constantly pushing the boundaries of what our platform can do.

And all of this contributes to something important: helping organisations identify and prevent financial crime before it causes harm. The sense of purpose adds an entirely different dimension to the engineering work.

Why You Might Want to Join Us

I joined as a mid-level engineer and became a senior not because I chased titles, but because the environment pushed me to think bigger and build better every single day. 

If, like me, you’re a software engineer who thrives on complexity, enjoys thinking about systems end-to-end and wants to work with people who genuinely care about their craft, this is a place where you’ll grow fast.

We’re hiring across multiple teams, and if any of this resonates with you, I’d love for you to check out our open roles on the ComplyAdvantage Careers Page.