Bricklayer AI

Provides a drag-and-drop interface for constructing multi-step data pipelines without code, using a node-based graph architecture where each node represents a data transformation, API call, or conditional branch. The builder compiles visual workflows into executable automation tasks that can be scheduled or triggered by webhooks, eliminating the need for traditional scripting in workflow orchestration.

Provides pre-built connectors to Bloomberg Terminal, Reuters, and academic financial databases with authentication handling and real-time data streaming capabilities. These connectors abstract away API complexity and handle rate limiting, data normalization, and credential management through a unified interface, allowing workflows to directly query market data without custom API code.

Accepts incoming data via webhook endpoints and processes it through workflows in near-real-time (latency <1 second). Webhooks support multiple authentication methods (API key, OAuth, HMAC signature verification) and can be configured to retry failed deliveries with exponential backoff. Workflows triggered by webhooks can emit their own webhooks to downstream systems, enabling event-driven architectures.

Executes automation workflows on a configurable schedule (cron-based intervals) or in response to external events via webhook endpoints. The execution engine maintains a task queue, handles retries with exponential backoff, and provides execution logs with step-by-step debugging information. Workflows can be paused, resumed, or manually triggered through the UI or API.

Provides a visual data mapper that transforms input data structures to output schemas through field-level mapping, type conversion, and expression-based transformations. Supports conditional field inclusion, array flattening, and nested object restructuring. The mapper generates transformation code (JavaScript or Python) that can be inspected and edited for advanced use cases, bridging visual and code-based approaches.

Provides step-level error catching with configurable retry policies, fallback paths, and alerting. Failed workflow executions are logged with full context (input data, error message, step where failure occurred), and alerts can be sent via email, Slack, or webhook. The monitoring dashboard displays workflow health metrics including success rate, average execution time, and failure trends over time.

Allows workflows to branch based on data conditions using if-then-else logic, with support for multiple conditions (AND/OR), comparison operators, and regex pattern matching. Branches can be nested and combined with loops to iterate over array data. The conditional engine evaluates expressions at runtime and routes execution to the appropriate branch, enabling dynamic workflow behavior based on data content.

Maintains workflow version history with the ability to revert to previous versions, though changes are not branched — only a linear history is maintained. Workflows can be exported as JSON for backup or sharing, and imported into other Bricklayer accounts. Deployment is immediate upon saving; there is no staging environment or approval workflow for production changes.

Exposes REST API endpoints to create, read, update, and delete workflows programmatically, as well as trigger workflow execution and retrieve execution history. API authentication uses API keys scoped to specific workflows or accounts. This enables external systems to manage Bricklayer workflows as code and integrate workflow execution into larger automation platforms.

Combines data from multiple sources (APIs, databases, files) into a single dataset with automatic deduplication based on configurable key fields. The aggregation engine handles schema mismatches by mapping source fields to a unified schema, and can apply transformations during aggregation (e.g., unit conversion, date normalization). Duplicate detection uses exact matching on key fields; fuzzy matching is not supported.

Automatically logs all workflow executions with timestamps, input data (if enabled), output data, user who triggered the workflow, and execution status. Logs are immutable and retained for a configurable period (default 30 days). Audit logs can be exported for compliance reporting and include metadata for regulatory requirements (e.g., data lineage, transformation history). Access to logs is role-based, with read-only access for auditors.