Castellum.AI

1. If your procurement process requires abundant third-party reviews and independent analyst validation: public review-platform coverage for the vendor is limited in the reviewed sources.
2. If you require transparent, published pricing and self-serve purchase options: the vendor’s public pricing page does not list specific dollar amounts in the reviewed crawl.
3. If your organization requires well-documented, publicly verifiable compliance certifications (for example, SOC 2 Type II, ISO certifications, data residency options, and subprocessor lists) prior to procurement.
4. If your institution is not prepared to establish governance and controls for AI-driven adjudication: automated adjudication raises MRM, QA sampling, and accountability requirements that must be supported internally.

Differentiators supported by available evidence (vendor-presented):

• Integrated stack: the vendor describes a combined offering of data, screening, and automated adjudication.
• Frequent update cadence: vendor materials state updates every 5 minutes.
• Explainable outputs: the vendor documents narrative explanations and audit-log concepts intended to support review and traceability.
• Modular deployment: API-driven data and screening modules can be adopted independently or with the vendor’s adjudication agents.

Where competitors may have advantages:

• Incumbents and larger vendors typically have broader published compliance collateral, more public reviews, and more established procurement track records.
• Some competitors publish entry-level pricing that can simplify initial procurement and budgeting.

Third-party marketing analysis and direct searches did not surface accessible Castellum.AI profile pages with end-user reviews on G2 or TrustRadius during this research pass. A Gartner Peer Insights product page was not found in accessible sources.

Interpretation: the absence of mainstream review-platform coverage can indicate that the product is earlier-stage, is primarily relationship-sold to enterprise customers, or that customers in regulated roles are less likely to leave public reviews.

The vendor publishes customer testimonials on its site. Example from a vendor case study:

"Castellum.AI is one of the most nimble vendors I’ve ever worked with, and they care about your ideas. Unlike other providers where you submit a ticket and wait weeks for a response, the team is always readily available for assistance and technical support. They put out a great product that allows us to truly own the risk and the process." — Daniel Schneider, Director of Financial Crimes, BSA Officer, SVP, Lead Bank

Interpretation: the testimonial emphasizes support responsiveness and configurability. These are vendor-curated testimonials and should be balanced with independent references where possible.

Stated results:

• A reported reduction in false positives (vendor-reported improvement)
• Reported time savings on screening reviews
• Implementation described as completed within a short timeframe
• Reported improvement in audit posture/regulatory readiness

Direct quote included in the vendor case study:

"Castellum.AI has given us the ability to perform all best-practice screening in house for ourselves and on behalf of our fintech partners at a high level, with little overhead in terms of ongoing monitoring and integration." — Daniel Schneider, Lead Bank

Critical evaluation / methodology gaps:

• Baseline systems and measurement methodology are not fully disclosed (for example, details about replaced systems, alert definitions, or time-study methods are not specified).
• Results may depend on tuning, list coverage changes, or workflow redesign—unclear how much is attributable solely to the vendor’s matching or automation.

Castellum.AI is a financial crime compliance screening platform focused on AML/KYC workflows. It combines (1) proprietary/aggregated global risk data (sanctions, PEPs, adverse media, export controls, etc.), (2) a screening and matching engine (including a described "Jgram" matching algorithm), and (3) automated AI capabilities marketed as AI agents that can auto-adjudicate or resolve alerts and generate audit-ready narratives. The vendor states the product is intended to reduce false positives and shorten Level 1/Level 2 alert review time in screening and monitoring workflows.

Key product areas described on the vendor site include sanctions & PEP screening, transaction screening, adverse media, KYC/KYB onboarding, real-time monitoring, and "Arbiter" AI agents for alert resolution. The vendor documents modular deployment options (for example, using their data + screening + agents, or layering agents over an existing screening provider).

Primary buyers/users: AML compliance, sanctions compliance, KYC operations, and financial crime teams at:

• Banks and credit unions
• Sponsor banks / BaaS banks
• Fintechs
• Crypto/digital asset firms
• Large enterprises with global supply chain / third-party screening needs

Positioning themes presented by the vendor include very frequent updates (vendor claims), explainable outputs intended to support auditability, and reduced false positive rates compared to legacy vendors (these are vendor claims and should be validated in procurement testing).