• OPTIMAL
  • BALANCED
  • VALIDATED
  • HUMAN

How to select balanced accelerator cohorts with your AI

Connect your AI to your candidate data once. After that, you can use it to build a cohort that hits every specific requirement you set.

Ask AI about this page

Short answer

How can I use AI to pick a balanced accelerator cohort?

Your AI scans your candidate pool and runs mathematical optimizations to find the best group based on your sector and diversity rules. You get a list of startups that perfectly fits your program's needs.

The results you get.

What your selection looks like.

The AI processes your constraints to produce these specific outcomes.

OPTIMAL

The best cohort mix

The AI finds the specific group of startups that satisfies every sector and diversity constraint you provide. You get a list of the most suitable candidates.

BALANCED

Diversity targets met

The tool checks the selection against your demographic or geographic goals. It ensures your cohort isn't skewed too heavily toward one single category.

VALIDATED

Constraint verification

The AI confirms the chosen group actually meets your hard limits. It flags if your requirements are impossible to meet with the current pool.

HUMAN

Final selection review

The AI handles the math, but you make the final decision. You review the suggested cohort to account for the qualitative nuance the data might miss.

The workflow

What your AI does when the candidate data arrives.

The AI handles the complex math of multi-variable optimization so you don't have to manually filter through spreadsheets.

  1. Ingest candidates

    Your AI reads the raw data from your application tracking system or spreadsheet to understand the available pool.

    calculate_optimal_composition

  2. Apply constraints

    The AI maps your specific needs, like sector limits or diversity targets, against the candidate attributes.

    calculate_optimal_composition

  3. Run optimization

    The agent calculates the best possible combination of startups to ensure the cohort is balanced and meets all rules.

    calculate_optimal_composition

  4. Verify composition

    The AI double checks the math to ensure the resulting group doesn't violate any of your predefined thresholds.

    calculate_optimal_composition

Try it

Copy these to start.

Use these prompts to kick off the selection process.

Starting points

Swap out the bracketed text for your actual data sources and specific cohort goals.

Accelerator Cohort Optimizer Connector

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

Connector linkhttps://edge.vinkius.com/vk_preview_wgnVXQSIIiuKXGYEG9LPVuLs3tHzdGBgE8IvMxQM/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 — Accelerator Cohort Optimizer capabilities are ready to use.
Configuration · claude_desktop_config.jsonCopy
{
  "mcpServers": {
    "accelerator-cohort-optimizer-mcp": {
      "url": "https://edge.vinkius.com/vk_preview_wgnVXQSIIiuKXGYEG9LPVuLs3tHzdGBgE8IvMxQM/mcp"
    }
  }
}
Copy into chat04

  • Look at my candidate spreadsheet and find a cohort of 20 startups that has at least 30% female founders.

  • Using the applicants in my Google Sheet, pick 15 companies that are distributed across Fintech, Healthtech, and SaaS.

  • Analyze my candidate pool and suggest a cohort that maximizes geographic diversity across Europe and North America.

  • Check the current applicant list and tell me if it's even possible to build a cohort with 5 AI companies and 5 Climate tech companies.

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

Start here

Connect Accelerator Cohort Optimizer once, then ask.

Link the MCP to your preferred AI client. Your credentials stay encrypted, and you can start running cohort optimizations immediately.

Connect Accelerator Cohort Optimizer to your AI

FAQ

How this task behaves.

  • 01

    Can the AI automatically send acceptance emails?

    No. The AI only calculates the optimal group based on your constraints. You must handle all communications manually.

  • 02

    Does the AI change my original candidate data?

    No. The AI only reads your data to perform calculations. It cannot edit, delete, or overwrite your original spreadsheets or databases.

  • 03

    What happens if my constraints are impossible to meet?

    The AI will tell you that no valid cohort exists that meets all your requirements. It won't force a selection that breaks your rules.

  • 04

    Can I use this with my existing spreadsheet?

    Yes. Once connected, your AI can read the data directly from the files you provide.

  • 05

    How does it handle diversity requirements?

    It treats diversity as a mathematical constraint, selecting candidates that satisfy the specific percentages or counts you request.

  • More questions about Accelerator Cohort Optimizer? The Connector page answers them. See everything the Accelerator Cohort Optimizer Connector can do

Connect Accelerator Cohort Optimizer to Claude, Cursor, ChatGPT & more