Financial crime compliance has become one of the most demanding parts of modern finance. Banks, credit unions, fintech companies, and crypto platforms all need to screen customers, monitor transactions, review alerts, and prove that their decisions are backed by clear evidence. The challenge is that many compliance teams are still working with slow systems, messy data, and too many false positives.
That is the problem Peter Piatetsky set out to solve with Castellum.AI.
As the Co-Founder and CEO of Castellum.AI, Peter has built a company focused on one of the most serious pain points in financial services: helping compliance teams move faster without weakening their controls. The platform brings together global risk data, AML and KYC screening, sanctions monitoring, PEP checks, adverse media, transaction screening, and explainable AI agents.
Peter’s story stands out because he did not enter the compliance space from the outside. Before building Castellum.AI, he worked in areas connected to sanctions, anti-money laundering, terrorist financing, and banking compliance. That gave him a direct view of how financial crime programs work in the real world, not just how they look in policy documents.
The result is a founder story built around practical experience, better data, and a clear understanding of what compliance teams actually need.
Who is Peter Piatetsky
Peter Piatetsky is best known as the Co-Founder and CEO of Castellum.AI, a financial crime compliance company built for regulated institutions. His career has been closely tied to the world of financial crime risk, sanctions, AML, and banking operations.
Before launching Castellum.AI, Peter worked in roles where financial crime risk was not a theory. It was part of daily decision-making. His background includes experience connected to the U.S. Treasury Department, where sanctions and financial crime policy play a major role in protecting the financial system. He also gained banking compliance experience, which helped him understand the pressure facing compliance teams inside financial institutions.
That mix of government, policy, and banking experience became important when he started building Castellum.AI. Many founders build software after spotting a market opportunity. Peter’s path was different. He saw the operational problem from the inside.
He understood that compliance teams were not struggling because they lacked effort. They were struggling because the tools around them often created more work than they removed. Watchlist data could be incomplete or hard to use. Alerts could pile up quickly. Analysts could spend hours reviewing names that only looked risky because of weak matching or poor data quality.
That experience shaped the way Castellum.AI was built. Instead of creating another surface-level workflow tool, Peter focused on a deeper issue: the quality of the risk data and the speed of the review process.
The financial crime compliance problem Peter Piatetsky saw up close
Financial crime compliance is full of pressure. Institutions need to follow rules, protect themselves from bad actors, and serve legitimate customers without unnecessary delays. That balance is hard to maintain.
A customer may be flagged because their name looks similar to someone on a sanctions list. A transaction may trigger an alert even when the activity is normal. A business may require extra review because ownership information is unclear. In each case, a compliance analyst may need to investigate, document the decision, and make sure the institution can explain what happened later.
The issue is not just the number of alerts. It is the quality of the alerts.
Too many false positives slow down compliance teams and create frustration across the business. Customer onboarding takes longer. Analysts spend time on low-risk cases. Real threats can become harder to spot because the team is buried under noise.
For banks, credit unions, fintechs, and crypto companies, this creates a serious operational challenge. They need strong AML and KYC controls, but they also need fast onboarding and smooth customer experiences. Legacy systems often make that difficult.
This is where Peter’s insight became valuable. He saw that financial crime compliance needed more than automation. It needed better data, smarter screening, and systems that could explain their decisions in a way compliance teams and regulators could trust.
Why Castellum.AI was built around better risk data
One of the most important ideas behind Castellum.AI is that compliance technology is only as strong as the data behind it.
If the data is outdated, incomplete, poorly structured, or difficult to interpret, the screening results will suffer. That can lead to missed risks, unnecessary alerts, and more manual work for analysts.
Castellum.AI focuses on in-house global risk data, which is a major part of its identity as a platform. The company works across areas such as sanctions, PEPs, adverse media, watchlists, customer screening, counterparty risk, and transaction screening.
This data-first approach matters because compliance teams need to know exactly what they are reviewing. They need to understand whether a customer is connected to a sanctioned person, a politically exposed person, a high-risk entity, or a negative news event. They also need to know whether a potential match is real or just a similarity in name.
Better data helps reduce noise. It gives analysts more context. It also makes AI more useful because an AI system trained or powered by weak information will still produce weak outcomes.
For Peter, the path to faster compliance did not start with flashy AI features. It started with the foundation. Castellum.AI was built on the idea that clean, current, enriched risk data can make compliance screening more accurate from the start.
How Castellum.AI helps compliance teams move faster
The promise behind Castellum.AI is simple: help compliance teams do their work faster, with more confidence, and with less manual strain.
The platform supports several parts of the financial crime compliance workflow. These include customer onboarding, KYC and KYB checks, sanctions screening, PEP screening, adverse media screening, transaction screening, alert review, and ongoing monitoring.
For a financial institution, each of these steps matters. A bank needs to know who its customers are. A fintech needs to onboard users quickly while still checking risk. A credit union needs strong controls without building a huge compliance technology stack. A crypto company needs updated risk intelligence because financial crime patterns can move quickly across digital asset networks.
Castellum.AI helps by combining screening tools with global risk data and AI agents designed for compliance workflows. The goal is not to replace human judgment. In financial crime compliance, people still matter. Analysts need to make decisions, understand context, and escalate serious cases.
The stronger use of AI is in removing repetitive work. If AI agents can review common alert patterns, gather supporting context, apply policy rules, and prepare an audit-ready explanation, analysts can focus their time on the cases that truly need human attention.
That is the practical value of Castellum.AI. It is not just about moving faster for the sake of speed. It is about giving compliance teams more room to think clearly and act on real risk.
The role of AI agents in Castellum.AI’s growth
AI agents have become a major part of the conversation around financial crime compliance, but the best use cases are practical rather than dramatic.
In the context of Castellum.AI, AI agents can support AML and KYC teams by helping with alert resolution, case review, onboarding checks, pKYC monitoring, and transaction screening. The company’s Arbiter AI Agents are designed to review alerts and support decision-making inside regulated compliance workflows.
This matters because many compliance tasks follow a similar pattern. An analyst receives an alert, checks the data, compares it against internal policy, reviews customer or transaction context, decides whether the alert should be closed or escalated, and documents the reasoning.
AI agents can help with that process by summarizing the alert, collecting relevant details, applying the institution’s rules, and producing a clear explanation. This can save time, especially for Level 1 and Level 2 alert reviews where teams often deal with high volumes.
Still, the important point is that AI in compliance cannot be treated like a black box. It has to work within policies, escalation rules, documentation standards, and audit requirements. Castellum.AI has leaned into that reality by focusing on explainable, policy-trained, audit-ready agents.
That approach fits Peter’s background. He understands that compliance teams do not just need a faster answer. They need an answer they can defend.
Why explainability matters in financial crime compliance
In many industries, speed is the main selling point for AI. In financial crime compliance, speed alone is not enough.
A bank or fintech cannot simply say that a tool made a decision. It needs to show what data was reviewed, why the decision made sense, how it matched internal policy, and whether the case should have been escalated.
That is why explainability is so important.
Compliance teams need clear audit trails. Regulators, internal auditors, and risk leaders may all need to understand how a case was handled. If an alert is closed, the reasoning should be documented. If it is escalated, the institution needs to know what triggered the escalation.
This is one of the reasons Castellum.AI has focused on explainable AI agents. The platform is built for a world where automation must be controlled, documented, and aligned with risk appetite.
For Peter Piatetsky, this is a key part of building technology for regulated industries. Financial institutions are not looking for vague AI promises. They need systems that fit into real compliance programs, support governance, and make examiners more comfortable rather than more concerned.
That is what separates useful compliance AI from simple automation.
How Castellum.AI serves banks, credit unions, fintechs, and crypto companies
Castellum.AI serves institutions that face serious financial crime risk but may have different operational needs.
Banks often need large-scale AML, sanctions, and transaction screening. They handle many customer types, payment flows, and regulatory requirements. For them, a platform like Castellum.AI can help improve screening accuracy and reduce the load on compliance teams.
Credit unions face a different challenge. They need strong compliance controls, but many do not have the same technology budgets or staffing levels as large banks. A modern financial crime platform can help them improve onboarding, monitoring, and alert review without adding unnecessary complexity.
Fintech companies need speed. They compete on fast customer experiences, but they cannot ignore KYC, KYB, sanctions, AML, or adverse media checks. If onboarding is too slow, they lose customers. If controls are too weak, they create regulatory risk. Castellum.AI fits into that middle ground by helping teams move quickly while still checking risk.
Crypto companies also need updated risk intelligence. Digital asset businesses operate in an environment where sanctions, fraud, scams, money laundering, and cross-border risks can change quickly. Strong screening and monitoring tools are important for managing that exposure.
Across all of these sectors, the core need is similar. Compliance teams want fewer weak alerts, better context, faster resolution, and documentation they can trust.
The Series A milestone and what it says about Castellum.AI’s momentum
A major part of Castellum.AI’s recent growth story is its $8.5 million Series A round. The funding marked an important step for the company and showed that investors see real demand for AI-powered financial crime compliance.
The round was led by Curql, a credit union-focused fintech investment collective. Other investors connected to the round included names such as BTech Consortium, Framework Venture Partners, Spider Capital, Remarkable Ventures, and Cameron Ventures.
This funding matters because it reflects a larger shift in the market. Financial institutions are under pressure to improve AML and KYC programs, reduce alert overload, manage sanctions risk, and adopt AI responsibly. They do not want tools that only look modern. They want tools that solve daily operational problems.
For Peter Piatetsky, the Series A was more than a funding announcement. It showed that Castellum.AI had moved from a strong idea to a company with growing market relevance.
It also placed the company in the middle of a bigger trend: the rise of AI agents in compliance operations. As more institutions explore AI, the winners will likely be the companies that combine speed with governance, data quality, and explainability.
That is where Castellum.AI is trying to lead.
What makes Peter Piatetsky’s founder story different
Peter’s founder story is strong because it has clear founder-market fit.
He did not build Castellum.AI because financial crime compliance sounded like a trendy software category. He built it after seeing the pain points directly. He understood sanctions risk from a policy and enforcement perspective. He understood banking compliance from an operational perspective. He also understood that many legacy systems were not built for the pace and complexity of modern financial crime risk.
That combination gave him a sharper view of the market.
Many compliance products focus on dashboards, workflows, or broad automation. Those things can be useful, but Peter’s approach with Castellum.AI goes deeper. The company focuses on the data layer, screening accuracy, AI-supported alert review, and documentation.
That is important because the hardest part of compliance is not always moving a case from one queue to another. It is knowing whether the case is truly risky, understanding why, and making a decision that can stand up to review.
Peter’s success comes from building around that reality.
How Castellum.AI reflects the future of financial crime compliance
The future of financial crime compliance will likely be shaped by three forces: better data, smarter automation, and stronger governance.
Financial institutions need to screen more customers, review more alerts, and respond to changing risks faster than before. At the same time, regulators expect clear documentation and responsible use of technology. That creates a difficult balance.
Castellum.AI reflects where the industry is heading. Compliance teams want AI that can support real work, not just generate summaries. They want tools that reduce false positives, improve onboarding, monitor risk, and help analysts make better decisions.
The next generation of AML and KYC platforms will need to be explainable by design. They will need to show how decisions were made, where the data came from, and how internal policy was applied. They will also need to work with human analysts, not around them.
This is why Peter’s work with Castellum.AI is timely. Financial crime is becoming more complex, and compliance teams are being asked to do more with limited resources. A platform that combines enriched risk data, sanctions screening, PEP checks, adverse media, transaction screening, and AI agents can give those teams a more practical way forward.
Why Peter Piatetsky’s work with Castellum.AI matters now
Peter Piatetsky built Castellum.AI around a problem that has become harder to ignore. Compliance teams are under pressure to move faster, but they cannot afford to be careless. They need stronger controls, cleaner data, fewer false positives, and tools that help them explain every important decision.
That is why Castellum.AI has gained attention in the financial crime compliance space. It speaks to a real need inside banks, credit unions, fintechs, and crypto companies. These organizations want modern compliance technology, but they also need trust, transparency, and auditability.
Peter’s achievement is not just building another AI company. It is building a company that applies AI to one of the most sensitive and complex areas of financial services. By combining in-house risk data, AML and KYC screening, sanctions monitoring, adverse media, and explainable AI agents, Castellum.AI is helping compliance teams reduce noise and focus on real risk.
In a market where financial crime threats continue to change, that kind of practical innovation matters.