- LP PAYOUT
- LP PROFIT
- GP CARRY
- FINAL SPLIT
Short answer
How do I calculate fund waterfall distributions with my AI?
Give your exit proceeds and tier parameters to your AI client. It runs the math through a four-tier model to split proceeds between LPs and GPs. You get a clear breakdown of every payout stage.
The distribution outcomes.
Where the proceeds land.
The AI processes the exit data and delivers these specific results.
LP PAYOUT
Return of capital
The AI calculates the initial capital returned to LPs. This covers the first tier of the waterfall.
LP PROFIT
Preferred return
The model identifies the hurdle rate and calculates the preferred return for LPs before any carry kicks in.
GP CARRY
Carried interest
The AI determines the GP's share once the preferred return is satisfied. It applies the specific carry percentage for the tier.
FINAL SPLIT
Human review needed
The AI produces a final summary of the split. You must verify these numbers against your legal LPA before sending payments.
The workflow
What your AI does when the exit data arrives.
The AI takes raw exit numbers and runs them through the mathematical model.
Ingest exit data
Your AI reads the total proceeds and capital account balances from your provided files.
calculate_waterfall_distributionApply tier logic
The agent applies the four-tier logic, moving from capital return to preferred return and then to carry.
calculate_waterfall_distributionCalculate GP share
The AI isolates the carried interest amount based on the specific tier thresholds.
calculate_waterfall_distributionGenerate summary
The agent produces a structured breakdown of the total distribution for both LPs and GPs.
calculate_waterfall_distribution
Try it
Copy these to start.
Use these prompts to trigger the calculation immediately.
Starting points
Swap the bracketed text with your actual fund details like the total exit amount or the hurdle rate.
Accelerator Carry Distribution Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_OiXUn5fWjNy9mfhXFSExDcE4cgLBkT38YtFDazPG/mcpClaude Desktop
Follow the steps below to connect in seconds.
- 1In Claude Desktop, open Settings → Connectors.
- 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
- 3Click Add and start a new chat — Accelerator Carry Distribution capabilities are ready to use.
{
"mcpServers": {
"accelerator-carry-distribution-mcp": {
"url": "https://edge.vinkius.com/vk_preview_OiXUn5fWjNy9mfhXFSExDcE4cgLBkT38YtFDazPG/mcp"
}
}
}Calculate the waterfall distribution for an exit of $50,000,000 using my fund's spreadsheet.
Run a four-tier waterfall calculation based on the capital accounts in my current project folder.
Show me the GP carry breakdown for a $10,000,000 distribution with an 8% preferred return.
Compare the LP and GP splits for the exit proceeds listed in my recent CSV upload.
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect Accelerator Carry Distribution once, then ask.
You only need to link the MCP to your client once. Your credentials stay encrypted, and you can start running calculations right away.
Connect Accelerator Carry Distribution to your AIFAQ
How the distribution task behaves.
- 01
Can the AI change my legal documents?
No. The AI only performs calculations based on the data you provide.
- 02
Does the AI handle clawbacks?
No. The current tool is designed for exit distributions, not retrospective clawback calculations.
- 03
Can I use this with Claude?
Yes, as long as you are using an MCP-compatible client.
- 04
Does the AI store my fund's private data?
No. Vinkius manages the connection, but the data is used only to perform the requested math.
- 05
Can the AI sign off on these distributions?
No. The AI provides the math, but a human must review and authorize all final payouts.
More questions about Accelerator Carry Distribution? The Connector page answers them. See everything the Accelerator Carry Distribution Connector can do
Connect Accelerator Carry Distribution to Claude, Cursor, ChatGPT & more
