# Investment thesis match screening with Eevo

> Eevo docs. Every page as Markdown is listed at https://eevo.io/llms.txt; the API is described at https://eevo.io/api/openapi.json.

The high cost/performance and speed/performance of decision models make them useful for many use cases that require processing large amounts of data, such as filtering search results by arbitrary natural language criteria.

There are many examples of this; one is screening investments for whether they match a given investment thesis. The high-level pattern for such use cases is:

1. The investor queries a database (like PitchBook or FactSet) with criteria based on readily existing structured data fields in the database, such as company size, geography and industry. This returns a list of, for example, ~500 initial results.
2. For each result, the investor uses a decision model to evaluate whether it matches their investment thesis, using natural language descriptions of the investment thesis and the company as input.

The code snippet below is an example of running such a query for a single company:

**Python**

```python
import os

import requests

THESIS = """A manufacturer of lighting equipment based in Germany, Austria or Switzerland that designs and makes its own products. Not a distributor or wholesaler, and not a contract manufacturer. Founder-led or family-owned, with no private equity backing. At most 500 employees, ideally under 100."""

COMPANY = """Lichtwerk Brenner GmbH, Linz, Austria
Founded in 1987 by Josef Brenner and run today by his daughter Anna, Lichtwerk Brenner is a family-owned company that designs and manufactures LED luminaires for offices, schools and hospitals at its own plant in Linz. Around 180 people work in design, production and service. Our luminaires are sold through electrical wholesalers and lighting planners across Austria, Germany and Switzerland."""

QUESTIONS = {
    "fit": {
        "type": "noul",
        "instructions": "Does this company fit the investment thesis?",
    },
    "under_100_employees": {
        "type": "noul",
        "instructions": "The company has fewer than 100 employees.",
    },
    "family_or_founder_owned": {
        "type": "noul",
        "instructions": "The company is founder-led or family-owned.",
    },
}

response = requests.post(
    "https://eevo.io/api/v1/systemone",
    headers={"Authorization": f"Bearer {os.environ['EEVO_API_KEY']}"},
    json={"model": "eevo-1", "state": f"Investment thesis\n{THESIS}\n\n{COMPANY}", "questions": QUESTIONS},
)
response.raise_for_status()
print(response.json()["answers"])
```

**TypeScript**

```typescript
const THESIS = `A manufacturer of lighting equipment based in Germany, Austria or Switzerland that designs and makes its own products. Not a distributor or wholesaler, and not a contract manufacturer. Founder-led or family-owned, with no private equity backing. At most 500 employees, ideally under 100.`;

const COMPANY = `Lichtwerk Brenner GmbH, Linz, Austria
Founded in 1987 by Josef Brenner and run today by his daughter Anna, Lichtwerk Brenner is a family-owned company that designs and manufactures LED luminaires for offices, schools and hospitals at its own plant in Linz. Around 180 people work in design, production and service. Our luminaires are sold through electrical wholesalers and lighting planners across Austria, Germany and Switzerland.`;

const QUESTIONS = {
  "fit": {
    "type": "noul",
    "instructions": "Does this company fit the investment thesis?"
  },
  "under_100_employees": {
    "type": "noul",
    "instructions": "The company has fewer than 100 employees."
  },
  "family_or_founder_owned": {
    "type": "noul",
    "instructions": "The company is founder-led or family-owned."
  }
};

const response = await fetch('https://eevo.io/api/v1/systemone', {
  method: 'POST',
  headers: {
    Authorization: `Bearer ${process.env.EEVO_API_KEY}`,
    'Content-Type': 'application/json',
  },
  body: JSON.stringify({ model: 'eevo-1', state: `Investment thesis\n${THESIS}\n\n${COMPANY}`, questions: QUESTIONS }),
});
if (!response.ok) throw new Error(`Eevo answered ${response.status}`);
console.log((await response.json()).answers);
```

**JavaScript**

```javascript
const THESIS = `A manufacturer of lighting equipment based in Germany, Austria or Switzerland that designs and makes its own products. Not a distributor or wholesaler, and not a contract manufacturer. Founder-led or family-owned, with no private equity backing. At most 500 employees, ideally under 100.`;

const COMPANY = `Lichtwerk Brenner GmbH, Linz, Austria
Founded in 1987 by Josef Brenner and run today by his daughter Anna, Lichtwerk Brenner is a family-owned company that designs and manufactures LED luminaires for offices, schools and hospitals at its own plant in Linz. Around 180 people work in design, production and service. Our luminaires are sold through electrical wholesalers and lighting planners across Austria, Germany and Switzerland.`;

const QUESTIONS = {
  "fit": {
    "type": "noul",
    "instructions": "Does this company fit the investment thesis?"
  },
  "under_100_employees": {
    "type": "noul",
    "instructions": "The company has fewer than 100 employees."
  },
  "family_or_founder_owned": {
    "type": "noul",
    "instructions": "The company is founder-led or family-owned."
  }
};

const response = await fetch('https://eevo.io/api/v1/systemone', {
  method: 'POST',
  headers: {
    Authorization: `Bearer ${process.env.EEVO_API_KEY}`,
    'Content-Type': 'application/json',
  },
  body: JSON.stringify({ model: 'eevo-1', state: `Investment thesis\n${THESIS}\n\n${COMPANY}`, questions: QUESTIONS }),
});
if (!response.ok) throw new Error(`Eevo answered ${response.status}`);
console.log((await response.json()).answers);
```

**Rust**

```rust
use serde_json::{json, Value};

const THESIS: &str = r#"A manufacturer of lighting equipment based in Germany, Austria or Switzerland that designs and makes its own products. Not a distributor or wholesaler, and not a contract manufacturer. Founder-led or family-owned, with no private equity backing. At most 500 employees, ideally under 100."#;

const COMPANY: &str = r#"Lichtwerk Brenner GmbH, Linz, Austria
Founded in 1987 by Josef Brenner and run today by his daughter Anna, Lichtwerk Brenner is a family-owned company that designs and manufactures LED luminaires for offices, schools and hospitals at its own plant in Linz. Around 180 people work in design, production and service. Our luminaires are sold through electrical wholesalers and lighting planners across Austria, Germany and Switzerland."#;

#[tokio::main]
async fn main() -> Result<(), Box<dyn std::error::Error>> {
    let questions = json!({
      "fit": {
        "type": "noul",
        "instructions": "Does this company fit the investment thesis?"
      },
      "under_100_employees": {
        "type": "noul",
        "instructions": "The company has fewer than 100 employees."
      },
      "family_or_founder_owned": {
        "type": "noul",
        "instructions": "The company is founder-led or family-owned."
      }
    });
    let response: Value = reqwest::Client::new()
        .post("https://eevo.io/api/v1/systemone")
        .bearer_auth(std::env::var("EEVO_API_KEY")?)
        .json(&json!({ "model": "eevo-1", "state": format!("Investment thesis\n{THESIS}\n\n{COMPANY}"), "questions": questions }))
        .send()
        .await?
        .error_for_status()?
        .json()
        .await?;
    println!("{}", response["answers"]);
    Ok(())
}
```

This would provide an answer like this:

```json
{
  "fit": {
    "type": "noul",
    "noul": 0.9206
  },
  "under_100_employees": {
    "type": "noul",
    "noul": 0.0486
  },
  "family_or_founder_owned": {
    "type": "noul",
    "noul": 0.9637
  }
}
```

The result provides probabilities of whether the company fits various aspects of the investment thesis. Naturally, the evaluated aspects are just an example and can be freely varied.

The resulting probabilities can be used to decide whether to analyze a company further, for example with a generative model or by people.

## Read next

- [Authentication](https://eevo.io/api/docs/reference/authentication.md): Create an API key and send it with each request.
- [Models, pricing and limits](https://eevo.io/api/docs/reference/models-pricing.md): Reference on Eevo models, pricing and request limits.
- [SystemOne decision](https://eevo.io/api/docs/reference/api-systemone.md): POST /api/v1/systemone
- [Data controls](https://eevo.io/api/docs/reference/data-controls.md): Controlling retention, training use and processing region of your requests.
