Skip to the content.

Mimic Studio User Guide

Mimic Studio is the visual IDE for OpenMimic, built with React 18, Vite, and Mantine. It enables interactive design of data generation streams, live previews, statistical quality checks, DAG stream topologies, and real-time observability.

Mimic Studio Live UI Screenshot


🎨 Studio Navigation & Feature Tabs

Mimic Studio provides 9 integrated workspaces accessible from the left sidebar:

  1. Generator Designer: Build, edit, and configure schemas, distributions, and sinks (Kafka, PostgreSQL, Redis, ClickHouse, S3, Stdout).
  2. Mimic Copilot: Interact with the natural-language generation assistant to convert plain English into validated Protocol Buffers configurations.
  3. Constraints & Solvers: Define field and cross-field predicates (amount > 50000 && fraud_flag == 1). The engine evaluates candidates and resamples until all predicates are satisfied.
  4. Quality & Fidelity: Inspect statistical fidelity metrics against reference data, including the two-sample Kolmogorov-Smirnov test (KS test) and Pearson correlation coefficient.
  5. Privacy & PII Scanner: Proactively detect unmasked sensitive identifiers (email, ssn, phone, card) and verify deterministic masking transforms.
  6. Starter Template Library: One-click launch production-grade blueprints (e.g. E-commerce Order Flow, Financial Fraud Stream, User Funnels).
  7. Audit Trail: Review immutable enterprise logs tracking generator starts, stops, and schema edits.
  8. Stream Topology & DAG: Orchestrate multi-stream directed acyclic graphs (e.g. Users stream emitting IDs before child Orders stream executes).
  9. Observability Dashboard: Real-time throughput (events/sec), latency p99, and error dashboards powered by Recharts.

🛠️ Step-by-Step Workflow

1. Designing a Generator

  1. Navigate to Generator Designer.
  2. Specify the Generator Name, Seed, and Event Limit.
  3. Choose your destination sink:
    • Kafka: Set topic (e.g. user.signups) and broker endpoints.
    • PostgreSQL: Configure connection string and destination table.
    • Redis: Choose key pattern and write mode (SET, LPUSH, RPUSH, PUBLISH, or XADD streams).
    • ClickHouse: High-throughput columnar table ingestion formatted as JSONEachRow.
    • Stdout: Formatted or raw NDJSON streaming directly to the terminal.
  4. Click Add Field to introduce new properties:
    • uuid: Deterministic RFC 4122 v4 UUID.
    • distribution: Normal or Uniform distribution for quantitative variables (e.g. latency, risk scores, payment amounts).
    • now: Logical timestamp formatted as RFC3339, epoch_ms, or epoch_s.
    • string_faker: Contextual strings (emails, names, locations, custom regex patterns).
    • weighted_choice: Probabilistic branches (success vs failure ratios).
    • entity_pool: Relational foreign-key references to guarantee customer/device entity reuse.
    • nested_object: Deeply nested JSON object structures.
    • array: Repeating variable-length collections.
    • media_meta: Image, audio, and video mock metadata with checksums and storage URIs.

2. Live Preview Streaming

Before streaming live events into production message queues or databases, use the Preview (10) feature:

Live Mock Data Preview with 10 Generated Records


3. Business Predicates & Constraint Solving

Switch to Constraints & Solvers:


4. Running & Stopping Pipelines


5. Multi-Generator DAG Orchestration

Switch to Stream Topology & DAG:


6. Observability & Throughput Dashboard

Switch to Observability Dashboard:

Real-Time Observability & Cluster Worker Dashboard


7. Geo Map & Hotspots

Switch to Geo Map & Hotspots:


8. Schema Version Control & Visual Diff

Switch to Schema Version Control:


9. Synthetic Camera Video Broadcaster

Switch to Synthetic Video Broadcaster: