• SETUP
  • DATA
  • AUDIT
  • HUMAN

How to manage ML datasets and projects with your AI

Connect Abacus AI to your agent to handle workspace organization through natural language. You'll move from high-level project planning to granular data inspection without touching the dashboard.

Ask AI about this page

Short answer

How can I manage ML datasets and projects with my AI?

You tell your agent to create new projects or check the details of specific datasets. It handles the workspace setup and metadata retrieval directly in Abacus AI. You get a clean, organized environment for your machine learning workflows.

Workspace outcomes

Where your work lands

What happens when you run these tasks.

SETUP

New project creation

Your agent builds out the project structure you request. This keeps your experiments isolated and organized from the start.

DATA

Dataset inspection

The AI pulls specific metadata for any dataset you point to. You get the technical details you need to validate your data before training.

AUDIT

Project inventory

Your agent lists every project currently in your workspace. This makes it easy to find where specific models or data reside.

HUMAN

Manual data uploads

The AI cannot physically upload raw files from your local machine. You still handle the actual data ingestion through the Abacus interface.

The workflow

What your AI does when the request arrives.

The agent acts as a bridge between your chat interface and the Abacus AI cloud.

  1. Inventory check

    The agent scans your current workspace to see what is already running. This prevents duplicate work and helps it understand your existing setup.

    list_projects

  2. Project initialization

    When you need a fresh environment, the agent builds the project container for you. This sets the stage for your upcoming ML tasks.

    create_project

  3. Data provisioning

    The agent sets up new datasets within your chosen project. This prepares the storage and schema for your training runs.

    create_dataset

  4. Metadata retrieval

    The agent looks up the specific properties and schema of a dataset. You get the technical context needed to verify data integrity.

    describe_dataset

Try it

Copy these to start.

Use these prompts to kick off your first task.

Starting points

Swap out the names in quotes for your actual project or dataset names.

Abacus AI Connector

You're all set. Choose your MCP client and follow the setup instructions.

Connector linkhttps://edge.vinkius.com/vk_preview_Lr50dCQPDxtMBNSIYHpuANfOgvy9zpJYnqAO4ozi/mcp

Claude Desktop

Follow the steps below to connect in seconds.

  1. 1In Claude Desktop, open Settings → Connectors.
  2. 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
  3. 3Click Add and start a new chat — Abacus AI capabilities are ready to use.
Configuration · claude_desktop_config.jsonCopy
{
  "mcpServers": {
    "abacus-ai-enterprise-ai-cloud-mcp": {
      "url": "https://edge.vinkius.com/vk_preview_Lr50dCQPDxtMBNSIYHpuANfOgvy9zpJYnqAO4ozi/mcp"
    }
  }
}
Copy into chat04

  • List all my current projects in Abacus AI.

  • Create a new project called 'Customer Churn Model'.

  • Create a dataset named 'user_logs_v1' in my 'Customer Churn Model' project.

  • Show me the metadata for the 'user_logs_v1' dataset.

  • Claude
  • ChatGPT
  • Cursor
  • VS Code
  • Windsurf
  • Claude Code
  • JetBrains
  • Cline

Start here

Connect Abacus AI once, then ask.

Just follow the link to link your account. Your credentials stay encrypted and secure on our side, and you can start chatting with your data immediately.

Connect Abacus AI to your AI

FAQ

How this task behaves

  • 01

    Can the AI delete my existing projects?

    No. The current tools only allow for listing and creating new projects.

  • 02

    Can I upload a CSV file directly through the chat?

    No. You must upload your files through the Abacus AI web interface; the AI can only create the dataset containers.

  • 03

    Does the AI see my training logs?

    No. The agent can only inspect dataset metadata and manage project structures.

  • 04

    Can the AI create a dataset in a different project?

    Yes, as long as you specify the target project name in your prompt.

  • 05

    Will the AI change my model hyperparameters?

    No. This connector is strictly for workspace and dataset management.

  • More questions about Abacus AI? The Connector page answers them. See everything the Abacus AI Connector can do

Connect Abacus AI to Claude, Cursor, ChatGPT & more