Modernising AML

Financial Crime Technology, and the Future of AML

The latest episode of FinTech Focus TV, recorded live at Financial Crime 360, brings together Toby and special guest Shani Golov, VP of Sales – EMEA at ThetaRay, for a compelling conversation about the evolution of financial crime prevention, the role of AI in AML, and the mindset shift needed across compliance teams to keep pace with fast-moving threats. 

Financial Crime Technology and the Purpose Driving Today’s FinTech Leaders

Toby opens the conversation by introducing Shani, who is preparing to speak on a later panel at the event. Shani explains her role at ThetaRay, where she manages the EMEA sales team and works closely with financial institutions adopting the company’s full suite of AML technologies. As she describes the offering, she outlines a complete end-to-end AML solution: onboarding screening, sanctions and adverse media checks, ongoing risk assessment, transaction monitoring, transaction screening, fraud detection, and long-term customer lifecycle analysis.

For Shani, the value of the product goes beyond technology; it ties into purpose. She speaks passionately about waking up each morning “with a big smile,” knowing she is helping fight criminals and supporting institutions in detecting cases “no one else can detect.” She emphasises that ThetaRay’s mission is grounded in uncovering the “unknown unknowns,” a powerful phrase that becomes a theme throughout the episode.

This sense of purpose resonates strongly across the financial crime community, and Toby reflects on this, noting that many people within AML and fraud prevention feel they play a noble role in protecting society. One previous guest even called themselves the “Batman of financial crime,” a comparison that underscores the importance and urgency of the work.

AI in AML and the Urgent Need for Mindset Shifts in Financial Services

As FinTech recruitment specialists, Harrington Starr has seen the rapid surge of demand for compliance talent, risk professionals, AML specialists, and data-driven technology teams. This episode highlights one of the biggest challenges facing these professionals: the mindset shift required to move beyond legacy rules-based compliance systems.

When Toby asks Shani what she hopes the audience will take away from her upcoming panel on modernising AML, her answer is direct: compliance teams must evolve their thinking. Many institutions remain comfortable with rules-based engines because those systems feel familiar and easy to explain to regulators. But as Shani points out, criminals adapt constantly, and traditional tools simply cannot keep up.

Adopting AI is not a quick switch; “it’s a journey,” she explains. It requires trust in the system, understanding of the explainability behind AI-driven recommendations, and a willingness to be proactive rather than reactive. Historically, compliance has been treated as a cost centre, a barrier to business rather than an enabler. Shani believes this perception must change. If financial institutions embrace advanced technology, compliance teams can become strategic drivers of growth rather than operational bottlenecks.

This shift mirrors what has happened in technology leadership roles more broadly. Toby notes that 25 years ago, CTOs weren’t widely recognised titles; IT leaders often lacked strategic influence. But as the industry adapted to increasing complexity, technical leadership moved to the forefront. Shani’s vision for the future of compliance reflects a similar transformation, one where AI and human expertise combine to elevate compliance from a defensive function to a proactive, value-adding capability. 

Machine Learning in Financial Crime: Understanding AI on Top, Supervised, and Unsupervised Models

One of the most detailed and informative segments of the episode comes when Shani breaks down the different types of AI used for transaction monitoring, clarifying a space where terminology is often used loosely across FinTech, RegTech, and AML recruitment conversations.

She outlines three core approaches:

AI on top of rules-based systems
In this setup, the core detection engine still relies on traditional rules. AI simply prioritises or classifies alerts, helping reduce some noise but still failing to detect unknown patterns. Institutions using this model still face high false-positive rates because the underlying mechanism is unchanged.

Supervised machine learning
Here, compliance teams “supervise” the system, defining what good and bad behaviour looks like. The system continually learns from analyst feedback, for example, marking alerts as true positives or false positives. But because humans cannot define patterns they do not yet know, supervised learning remains limited and biased by existing knowledge. Shani describes this approach as “AI at the investigation level”: helpful for triage, but not transformative for detection.

Unsupervised machine learning, ThetaRay’s differentiator
This is where ThetaRay stands out. Their unsupervised machine learning system detects anomalies without needing pre-defined rules or labels. It learns each customer’s behaviour based on large historical datasets, identifying what is normal for one person versus another. When something deviates significantly from expected behaviour, the system flags it as suspicious.

Shani emphasises that knowing what you don’t know is at the heart of effective financial crime detection. Criminals use AI to become more sophisticated, agile, and evasive, and only equally advanced AI can match and surpass them.

She clarifies that ThetaRay’s solution blends unsupervised learning, supervised feedback looping, and rules to provide the strongest, most accurate detection engine possible.

This hybrid, multi-layered approach is increasingly sought after in the FinTech recruitment space, where financial crime teams are rapidly hiring data scientists, ML engineers, AML product specialists, and compliance technologists capable of working with multidimensional AI-driven tools.

Human Insight + AI Automation: The Future of AML Talent in Financial Services

Another key message throughout the episode reinforces something Harrington Starr sees regularly when placing compliance and AML professionals: AI will not replace analysts. Instead, it will elevate them.

Toby mentions that many people across the industry feel uneasy about whether AI could diminish the importance of human expertise. Shani counters this clearly and confidently: AI makes analysts faster and better, but the final decisions will always require human judgment.

In her words, AI “helps the analyst to take faster and better decisions,” but “it’s not going to replace the human being.”

This balance is especially important in financial services recruitment, where institutions want compliance teams who can manage intelligent systems, interpret complex outputs, and collaborate with technology functions. The future AML workforce must be both tech-savvy and business-savvy, comfortable bridging the gap between engineering and regulation.

What 2026 Holds for AML Technology and FinTech Talent

When Toby asks Shani to look ahead to 2026, she predicts the increasing rise of virtual agents, automated systems capable of conducting entire investigations and presenting comprehensive summaries to analysts.

These virtual agents will not make final decisions. Rather, they will collect and structure information, highlight the most relevant points, and even recommend potential conclusions. But the ultimate approval will remain with compliance teams.

Toby draws a parallel to the changes he has seen across the recruitment industry, the evolution from manual processes to tech-enhanced workflows. He identifies three words that define the future of AML work: “faster, better, stronger.” Technologies like ThetaRay’s enable teams to be more efficient, more accurate, and more productive.

Shani agrees that efficiency and productivity will be among the most important themes shaping financial crime teams over the next few years. The pressure on institutions to detect threats earlier, reduce false positives, and manage increasingly sophisticated criminal networks means that technology and talent must evolve in parallel. 

ThetaRay’s Mission, Market Strength, and Commitment to Innovation

Throughout the conversation, Shani repeatedly returns to the principles that guide ThetaRay’s product development: explainability, trust, sophistication, and the ability to detect previously unknown threats. She describes the company as having a unique blend of purpose-driven culture, mission-critical technology, and long-standing market experience, having operated successfully for more than 13 years.

Toby reflects on hearing from one of Shani’s colleagues earlier in the day, who expressed strong confidence in the quality and competitive advantage of the solution. Shani emphasises that the combination of unsupervised learning with supervised elements and traditional rules provides financial institutions with the most complete and effective AML technology stack available.

ThetaRay aims not simply to reduce workloads, but to detect the earliest signs of criminal activity, the anomalies other systems miss. This, Shani explains, is where true impact lies.

FinTech Recruitment, Compliance Hiring, and the Skills Needed for the Next Era of AML

As the financial crime landscape evolves rapidly, this conversation underscores the importance of strategic hiring within compliance, financial crime, AI engineering, and FinTech product teams. Harrington Starr continues to support institutions seeking talent capable of operating within this next generation of AML technology, from ML engineers to AML analysts, data experts, and RegTech sales leaders like Shani.

The industry is shifting toward solutions that require sophisticated interpretation, cross-functional collaboration, and comfort with AI-driven tools. The professionals who succeed in this space will be those who embrace technology, adapt quickly, and adopt the “mindset shift” Shani champions throughout the episode.