CredVault Intelligence Engine (CIE)
The CredVault Intelligence Engine (CIE) is a powerful, developer-first AI and Machine Learning platform. Rather than dealing with clunky web interfaces, CIE is driven entirely by a lightweight, lightning-fast Command-Line Interface (CLI) that is deeply integrated into your database, analytics, and robotics workflows.
Installing the CLI
To get started with CIE, install the official CLI tool via npm. This makes the cie command globally available on your terminal.
npm install -g credvault-cie
Once installed, authenticate your terminal session with your CredVault account. You can optionally scope your login to a specific enterprise workspace.
cie login --tenant <tenant-id>
1. The Machine Learning Lifecycle
CIE simplifies the entire end-to-end Machine Learning lifecycle into a few simple terminal commands.
Uploading Training Data
Prepare your datasets (CSV, JSON, etc.) and seamlessly stream them into your secure CredVault storage bucket.
cie data upload customers.csv
Training Models
Kick off a background training job on CredVault's high-performance compute clusters. You can specify the type of model (e.g., classification, regression) or let CIE auto-detect the best architecture.
cie train ds_abc123 --name "Customer Prediction Model" --type classification
You can monitor the status of your training job at any time:
cie job status job_xyz789
Deploying for Production
Once training is complete, deploy the model to a scalable API endpoint in milliseconds.
cie deploy md_123456 --env production
2. Advanced: Direct SQL Execution
Bypass NoSQL syntax completely when you need standard analytical power. The CLI possesses a native querying engine that automatically translates ANSI SQL queries against your underlying LakeVault clusters.
cie query "SELECT name, age FROM users WHERE active = true ORDER BY created_at DESC"
CIE will automatically execute the query securely and format the results in a beautiful ASCII table right in your terminal.
3. Advanced: The Vault AI Agent
CIE isn't just an interface; it has a conversational, reasoning AI embedded directly into the terminal, aware of your entire cloud environment.
Conversational Assistance
Stream thoughts, code snippets, and instructions directly from the Vault AI:
cie ask "How do I optimize my latest classification model?"
🩺 CIE Environment Doctor
If you are experiencing cluster issues or deployment failures, do not debug manually. Run the interactive Doctor tool.
cie doctor
The AI agent will perform a proactive diagnostic scan of your environment, read your logs, find the configuration errors, and interactively present you with the very CLI commands needed to fix them. You can select checkboxes in the terminal to let the AI fix your cluster automatically.
4. Advanced: IoT & Robotics Integration
CIE is natively optimized for edge deployments. If you are running physically separated robotics hardware, you can bind devices to your workspace and stream live sensor inference using WebSockets directly to the console.
Connect a device:
cie robot connect robot_001
Stream live telemetry (Battery, Temperature, CPU, GPS coordinates):
cie robot monitor robot_001
Sync your deployed ML models bi-directionally down to the edge hardware layer:
cie sync