Shield Connectors

Caju Export Connector

Back To Connectors Page

Mobile Messaging & Client Communications | Connector for Caju AI Export

Request a Demo

What Is the Shield Connector for Caju Export?

Caju AI is a mobile communications governance and compliance platform used by regulated financial services firms to enable, govern, and capture client conversations across consumer messaging channels, including WhatsApp, WeCom, SMS, Signal, Telegram, and voice, within a compliance-grade enterprise environment.

Caju AI enables regulated employees to communicate with clients through their preferred messaging applications via a Microsoft Teams integration or native Caju interface, while applying enterprise governance controls, including Intent-based DLP, information barriers, and real-time monitoring, at the platform layer. Caju Export refers to Caju AI’s data export capability: the API-based mechanism through which Caju AI exports captured communications data to third-party archiving and compliance platforms, including Shield.

For regulated financial services firms, Caju AI addresses the mobile messaging governance and capture challenge, enabling consumer messaging channels to be used for regulated client communications in a controlled, capturable environment. Caju AI’s own platform includes monitoring, alerting, and some compliance capability; however, the complete financial services compliance obligation requires integration with a dedicated AI-powered surveillance platform. Caju Export is the bridge that connects Caju’s governed capture environment to Shield’s compliance archive and surveillance engine.

Shield’s connector for Caju Export ingests communication records and associated metadata exported from the Caju AI platform directly into Shield’s compliance platform, making every captured client interaction available for AI-powered surveillance, investigation, and eDiscovery alongside every other channel the firm uses. From the moment data enters Shield, it is available in a single unified platform — without the surveillance gap that exists when mobile communications are governed but not surveilled.

Caju-exported data does not exist in isolation. Shield understands the full context of these communications, enabling compliance teams to detect genuine risk rather than chasing false positives.

Why Caju Export Compliance Is Complex

Caju AI-governed communications present specific compliance challenges that go beyond the technical capture question. Several issues arise consistently across regulated firms:

  • The gap between Caju AI governance and Shield surveillance. Caju AI provides mobile communications governance — applying DLP policies, information barriers, and compliance controls to mobile messaging channels, capturing communications, and providing monitoring and alerting capabilities. What Caju AI does not provide is the AI-powered financial services misconduct surveillance, cross-channel correlation, and regulatory eDiscovery workflows that SEC, FINRA, MiFID II, and FCA compliance frameworks require. Caju is the governance and capture layer; Shield is the compliance surveillance layer.
  • Multi-channel capture from a single export stream. Caju AI captures communications from multiple consumer messaging channels simultaneously — WhatsApp, WeCom, SMS, Signal, Telegram, and voice — through a single platform. The Caju Export data stream therefore contains communications from multiple channels, each with different data structures, metadata characteristics, and compliance risk profiles. Shield’s connector correctly handles this multi-channel export stream, preserving channel-specific metadata and applying appropriate surveillance models to each communication type.
  • Microsoft Teams integration and interface context. Caju AI integrates with Microsoft Teams, enabling employees to communicate with messaging-app-using clients directly from their native Teams interface. This means that Caju-exported communications may carry both Teams-level metadata (the interface through which the employee communicated) and the consumer messaging channel metadata (the channel through which the client received the message). Both layers are material to the compliance record.
  • AI-generated summaries and metadata. Caju AI applies generative AI capabilities to captured communications — generating call summaries, conversation insights, and flagging potential compliance concerns. The Caju Export stream may include both raw communication content and AI-generated metadata. Shield’s connector correctly handles both, preserving AI-generated context while applying Shield’s own financially-calibrated surveillance models to the underlying communication content.
  • Cross-channel continuity. Client interactions captured through Caju AI accompany Bloomberg IB, email, and voice communications about the same matters. Shield correlates all channels — including Caju-exported mobile communications — in a single investigation and examination workflow.

Key Features of the Shield Caju Export Connector

Complete Multi-Channel Caju Export Ingestion. Shield ingests all communications data exported from the Caju AI platform, including WhatsApp, WeCom, SMS, Signal, Telegram, and voice communications captured by Caju AI, preserving full message content, voice recordings, file attachments, and every metadata field from the Caju export stream. All data is ingested in full, with zero data loss across all Caju-captured communication channels.

Full Multi-Layer Metadata Preservation. Shield retains and enriches the complete Caju Export metadata layer — including communication channel identifiers, employee and client identifiers, conversation thread context, timestamps, message IDs, Caju governance policy context, Teams interface metadata where applicable, and AI-generated summary data from the Caju platform. All metadata layers are preserved in their original form, made fully searchable, and stored as part of the immutable compliance record.

AI-Powered Financial Services Surveillance. Applied to Caju Data Shield applies AI surveillance models specifically calibrated for financial services communications to Caju-exported data — targeting market manipulation, information leakage, MNPI sharing, front-running, and personal misconduct across all communication channels captured by Caju AI. This provides materially higher-quality financial services misconduct detection than general-purpose compliance monitoring, ensuring that risk signals in WhatsApp, WeChat, and SMS communications are identified accurately in the specific language patterns of regulated financial services.

Immutable, Audit-Ready Archive. All Caju Export data captured by Shield is stored in an immutable, tamper-evident archive with a full audit trail. Retention periods are configurable to meet jurisdiction-specific requirements — including the six-year standard under SEC Rules 17a-3 and 17a-4, the five-to-seven-year requirements under MiFID II and MAR, and the five-year minimum under FCA SYSC 10A.

Out-of-the-Box AI Surveillance Models. Shield ships with pre-configured AI surveillance models for Caju-exported communications, calibrated to the specific language patterns and risk typologies of mobile financial services communications across all channels supported by Caju AI. Models can be customised to reflect a firm’s specific risk appetite, restricted lists, and internal policy requirements.

Unified Cross-Channel Surveillance. Shield ingests Caju Export data into the same unified compliance platform as Bloomberg IB, email, Teams, and every other channel — enabling compliance teams to correlate Caju-captured mobile client communications with all other channel data in a single investigation and examination workflow.

Regulatory Coverage

Mobile communications data exported from Caju AI is classified as business records subject to capture, retention, and surveillance requirements across multiple regulatory frameworks:

  • SEC Rules 17a-3 and 17a-4 — Requiring broker-dealers to capture, preserve, and produce records of all business communications — applicable to Caju-governed mobile communications by registered representatives and regulated employees, stored in WORM-compliant format for a minimum of six years.
  • FINRA Rules 4511 and 3110 — Requiring member firms to archive all business communications including mobile messaging with written supervisory procedures in place — applicable to communications captured via Caju AI.
  • MiFID II Article 16(7) and Market Abuse Regulation (MAR) — Requiring investment firms to retain electronic communications related to orders and transactions for a minimum of five years — applicable to Caju-governed mobile client communications by EU-regulated firms.
  • FCA Rules (SYSC 10A and MAR) — Requiring FCA-regulated firms to retain relevant electronic communications for a minimum of five years and implement effective surveillance — applicable to Caju-governed mobile communications by regulated UK employees.
  • CFTC Regulation 1.35 and 17 CFR § 23.202 — Applicable to Caju-governed mobile communications about swap and commodity trading activity.
  • MAS, SFC/HKMA, ASIC, and equivalent APAC regulations — Applicable to firms with APAC regulatory obligations using Caju AI for client communications across APAC markets.
  • GDPR and applicable data privacy regulations — Shield’s architecture supports data residency requirements and privacy-compliant data handling across all jurisdictions in which Caju AI is deployed.

Other Related Connectors

Frequently Asked Questions

Which messaging channels does the Caju Export connector cover?

The Caju Export connector ingests communications from all channels captured by Caju AI and included in the Caju export stream — including WhatsApp, WeCom (WeChat enterprise), SMS, Signal, Telegram, and voice. The specific channels covered depend on which Caju AI modules are deployed at each firm. Shield’s connector handles the multi-channel export stream, preserving channel-specific metadata and applying appropriate surveillance models to each communication type.

Can Caju Export data be reviewed alongside Bloomberg IB and other channels during an investigation?

Yes. Shield ingests Caju Export data into the same unified compliance archive as Bloomberg IB, email, Teams, voice, and every other channel — enabling complete cross-channel client communication reconstruction in a single investigation workflow.

How does Shield handle the multi-channel nature of Caju Export data?

Caju AI captures communications from multiple consumer messaging channels simultaneously and exports them through a single API stream. Shield’s Caju Export connector correctly parses this multi-channel stream, preserving channel-specific metadata and applying the appropriate surveillance models and regulatory treatment for each communication type within the unified Shield compliance archive.

Related Terms