Jube

Overview

Jube is a self-hosted platform for detecting suspicious transactions and managing AML and fraud cases. It suits financial institutions and fintechs that want control over deployment and can take responsibility for operating the software. Its broad detection and investigation toolkit is compelling, though production use calls for clustered deployment planning and careful attention to AGPLv3.

Key features

Detection and explainability

Jube combines thresholds, velocity and aggregation checks, sanctions screening, and supervised and unsupervised machine learning. Its exhaustive adaptation approach tries neural-network structures and variables, giving teams several detection methods to bring together. That range can serve complex monitoring needs, but it also makes specialist calibration and validation important. The maker says Bayesian Network and decision-tree outputs can be explained to regulators and describes calibration and validation against BASEL practice, a useful focus for institutions that must account for decisions.

Investigation workflow

AML, fraud, and compliance cases share investigator dashboards, document versioning, automatic escalation, and audit trails. Carrying investigation tools alongside monitoring can help teams move from alert to review within one system, and the three streams suit organisations whose compliance work extends beyond payment fraud.

Interfaces and deployment

Synchronous and asynchronous interfaces run over HTTP, AMQP, or hybrid modes. Redis supports real-time state, PostgreSQL handles durable storage and audit logs, and RabbitMQ connects inbound and outbound events through AMQP. Docker and Kubernetes support container deployment, while multi-tenancy and the choice of an organisation’s infrastructure or cloud tenancy offer deployment flexibility. Keeping Jube in the customer’s infrastructure and jurisdiction can help retain data control, but self-hosting also means the organisation must plan its own production environment. The Docker Compose setup is for evaluation and development; production deployments should follow the clustered deployment documentation.

License and support

Jube is fully open source under AGPLv3, with all features available. Organisations that modify the core software and run it as a network-accessible service must make those modifications available to users of that service, a material consideration for teams adapting the platform.

Support is async-first over WhatsApp, with requests tracked as tickets and answered through Loom videos and documentation. That model suits teams comfortable resolving issues asynchronously. Enterprise support adds security patching for credible, documented vulnerabilities, prioritized by severity and exploitability, and a guaranteed 06:00–18:00 GMT Monday-to-Friday callout response; coverage outside those hours is best endeavours.

Pricing

The Jube software is free at no cost, with all features available under AGPLv3. There is no free trial because the software itself has a free plan. The paid monthly plans buy support and service coverage rather than additional software features.

• Sandbox support: $350 a month, for a non-production sandbox, with an async response within 48 hours. It fits evaluation and development, not a live pipeline.
• Growth support: $950 a month, for one live production pipeline under roughly 100,000 transactions a month, with next-business-day response. The single-pipeline and volume limits make it a defined entry point for smaller production operations.
• Scale support: $1,590 a month, for multi-region pipelines with same-day, high-priority routing. It suits deployments that need broader coverage and faster routing than Growth.
• Enterprise support: $2,400 a month, billed at $2,400 / mo ($28,800/yr), on a 12-month initial term. It covers mission-critical infrastructure, a 24/7 priority emergency SLA, and advanced analytics consulting, alongside the stated security support.

For training, public training in Kuala Lumpur costs $1,825, paid once per participant for three days, capped at six participants, and includes materials, workshops, meals and refreshments, plus up to six weeks of remote support. Private training starts at $6,000, paid once for three days; on-site delivery depends on location and scope.

Platforms

Jube supports API and web access, Linux, and self-hosted deployment. Docker and Kubernetes provide containerized deployment options.

Who it's for

Jube is best suited to compliance teams, financial institutions, and fintechs that need real-time AML or fraud detection together with case management, and have the infrastructure and expertise to operate a self-hosted platform. Organisations that need a managed service, or cannot meet the AGPLv3 conditions for modified network-accessible software, should look elsewhere.

Pros and cons

• Pros: Detection combines rule checks, sanctions screening, and several machine-learning approaches, letting teams build a varied monitoring strategy.
• Pros: Case workflows cover AML, fraud, and compliance with versioning, escalation, and audit trails in the same platform.
• Pros: Free access includes all features, and deployment on the organisation’s own infrastructure can help keep data within its control.
• Cons: Production use requires clustered deployment planning; the Docker Compose setup is only for evaluation and development.
• Cons: AGPLv3 requires modified core software run as a network-accessible service to be made available to users of that service.
• Cons: Growth support is limited to one live pipeline under roughly 100,000 transactions a month, while its support response is next business day.

Alternatives

Compare payment fraud detection software if you want to consider products across the category.

• FraudLabs Pro is a freemium option with a free trial and a Micro plan capped at 500 queries a month, sales under 25K USD a month, and five validation rules. Choose it when that query-based scope fits better than Jube’s self-hosted platform.
• IPQualityScore Fraud Prevention offers a free plan with 10 custom rules, 1,000 lookups a month, and 35 a day, plus a free trial. It may suit teams that want to start with a defined lookup allowance.
• Stripe Radar includes Radar Lite at no additional charge for businesses on standard payments pricing, with AI fraud prevention for card payments and card testing. Consider it when that card-payment focus and included pricing fit.
• Ojuri is a free, MIT-licensed option with no per-call fees or rate limits. It is an alternative for readers comparing free software under a different license.
• Ravelin offers a customized platform with pricing based on business goals, risk appetite, and setup; contact it for a demo or call.
• Tazama has no software or license charge, while hosting, implementation, and support require resources or a contract. It may fit teams prepared to provide those resources.
• Vesta is a paid fraud prevention alternative.
• DataVisor Fraud Platform is a custom enterprise platform with pricing on request; consider it if you want to explore an enterprise offering through a demo.

Verdict

Choose Jube if your institution needs configurable real-time transaction monitoring and AML case management in one self-hosted, open-source platform, and can handle its deployment and license obligations. Its strongest reason to choose is the combination of broad detection methods, explainable outputs, and investigation workflows without a software fee. Look elsewhere if your team needs a managed product, cannot operate clustered production infrastructure, or needs support coverage beyond the limits of its chosen tier.