Extruct.ai: Company Research API for AI Agents

Company research is still too manual for agents. A human can bounce between databases, search engines, LinkedIn, maps, and CRM records until the picture is clear. An agent needs that same context through APIs: discover the right companies, score them against a thesis, enrich the records, and carry source-backed answers into the next workflow.

Today we're excited to announce our partnership with Extruct.ai, a company research API for AI. Extruct helps agents find and research companies from a curated 10 million company index or the live web, with sources attached to the answers that come back.

What is Extruct.ai?

Extruct gives agents a practical company research stack. Semantic Search is the fast path for natural-language company discovery over the Extruct index. Lookalike Search starts from a representative domain or company UUID and returns similar companies. Deep Search runs asynchronous discovery tasks that evaluate companies against natural-language criteria using the web, Extruct DB, Maps, and LinkedIn. Deep Research produces cited reports on companies, people, or teams. AI Tables let teams enrich, score, and review company lists in reusable workflows.

That combination matters because company research rarely stops at one lookup. A sourcing agent may start from a market thesis, expand from a seed company, rank candidates against a custom rubric, enrich the shortlist, and hand the output to a CRM. A sales agent may need an account plan with current initiatives, buyers, vendors, and conversation angles. Extruct gives those workflows one place to discover, research, and structure company intelligence.

Using Extruct.ai with Orthogonal

Extruct.ai is available through the same Orthogonal SDK your agents already use. One Orthogonal key gives you access to Extruct alongside the rest of your research, GTM, and enrichment stack.

Search for Companies

Use natural language to find companies by category, use case, ICP, or problem space. The POST search endpoint keeps filters as JSON, which is easier for agents to compose and safer when exclusion lists grow.

import Orthogonal from "@orth/sdk";

const orthogonal = new Orthogonal({
  apiKey: process.env.ORTHOGONAL_API_KEY,
});

const result = await orthogonal.run({
  api: "extruct-ai",
  path: "/v1/companies/search",
  method: "POST",
  body: {
    q: "vertical SaaS for logistics procurement",
    filters: {
      include: {
        country: ["United States"],
        size: ["51-200", "201-500"]
      },
      exclude: {
        domains: ["stripe.com"]
      }
    },
    pagination: {
      offset: 0,
      limit: 20
    }
  }
});

console.log(result.data.results);

Look Up a Company Profile

Start with a known domain and get the canonical enriched Extruct profile. The response can include descriptions, products, use cases, audience, firmographics, social profiles, and sources when available.

const profile = await orthogonal.run({
  api: "extruct-ai",
  path: "/v1/companies/stripe.com",
  method: "GET",
});

console.log(profile.data.context);

Find Similar Companies

Use a strong seed company to expand into lookalikes. This is useful when an agent has one ideal customer, competitor, or portfolio company and needs more companies with a similar profile.

const similar = await orthogonal.run({
  api: "extruct-ai",
  path: "/v1/companies/rippling.com/similar",
  method: "GET",
  query: {
    offset: "0",
    limit: "20",
  },
});

console.log(similar.data.results);

Run Deep Search

When filters are not enough, create a discovery task that scores candidates against qualitative criteria. Extruct returns task progress and paginated results, including criterion-level grades, explanations, and sources.

const task = await orthogonal.run({
  api: "extruct-ai",
  path: "/v1/discovery_tasks",
  method: "POST",
  body: {
    query: "vertical SaaS companies serving freight forwarding",
    desired_num_results: 150,
    criteria: [
      {
        key: "has_logistics_focus",
        name: "Logistics Focus",
        criterion: "Company serves freight forwarding or logistics operations."
      },
      {
        key: "b2b_gtm",
        name: "B2B GTM Fit",
        criterion: "Company sells primarily to business buyers."
      }
    ]
  },
});

console.log(task.data.id);

Read Deep Search Results

Poll the task, then retrieve results once enough companies have been evaluated. Agents can use the explanations and sources to decide which candidates deserve enrichment, outreach, or human review.

const results = await orthogonal.run({
  api: "extruct-ai",
  path: `/v1/discovery_tasks/${task.data.id}/results`,
  method: "GET",
  query: {
    offset: "0",
    limit: "50",
  },
});

console.log(results.data.results);

Create a Deep Research Report

Deep Research turns a detailed brief into a cited report about a company, person, or team. Use it for account planning, buyer research, initiative summaries, diligence, and other workflows where a shallow profile is not enough.

const research = await orthogonal.run({
  api: "extruct-ai",
  path: "/v1/deep_research_tasks",
  method: "POST",
  body: {
    brief: "We sell cloud cost optimization software. Research Shell's IT and digital organization, recent cloud initiatives, likely infrastructure buyers, and practical conversation angles for a FinOps sales motion.",
    depth: "medium",
  },
});

console.log(research.data.id);

Using x402 Protocol

Extruct.ai on Orthogonal supports x402 for native HTTP payments. Agents can pay for individual API calls with USDC, no separate Extruct key required.

import { wrapFetchWithPayment } from "x402-fetch";
import { privateKeyToAccount } from "viem/accounts";

const account = privateKeyToAccount(
  process.env.PRIVATE_KEY as `0x${string}`
);
const fetchWithPayment = wrapFetchWithPayment(fetch, account);

const response = await fetchWithPayment(
  "https://x402.orthogonal.com/extruct-ai/v1/companies/stripe.com",
  { method: "GET" }
);

const data = await response.json();
console.log(data.context);

Using MPP

Extruct.ai also works through MPP, the open standard for machine-to-machine payments co-authored by Tempo and Stripe. Agents can pay for Extruct calls with stablecoins, cards, or Bitcoin on the same endpoint.

import { privateKeyToAccount } from "viem/accounts";
import { Mppx, tempo } from "mppx/client";

Mppx.create({
  methods: [tempo({ account: privateKeyToAccount("0x...") })],
});

const response = await fetch(
  "https://mpp.orthogonal.com/extruct-ai/v1/companies/search",
  {
    method: "POST",
    headers: { "Content-Type": "application/json" },
    body: JSON.stringify({
      q: "AI-native CRM tools for venture investors",
      pagination: { offset: 0, limit: 20 }
    })
  }
);

const data = await response.json();
console.log(data.results);

The mppx client handles the 402 challenge, signs a credential with your wallet, and retries with proof of payment.

What Agents Can Do with Extruct.ai

Agents can use Extruct to move from intent to structured company intelligence. A GTM agent can search for companies matching a narrow ICP, look up each company profile, score the shortlist with Deep Search, and enrich the final rows in AI Tables. A research agent can prepare an account plan before a meeting, including source-backed context about initiatives, teams, vendors, and likely priorities. A sourcing agent can start with one portfolio company, find lookalikes, rank them against a thesis, and keep a table refreshed as new evidence appears.

The differentiator is that Extruct is not only a company database. It combines indexed company discovery, live-web research, source attribution, asynchronous agent workflows, and repeatable table-based enrichment. That lets agents ask higher-level research questions while still getting structured results they can pass downstream.

Why We Partnered with Extruct.ai

Agents need company research that behaves like infrastructure. It has to be searchable, source-backed, structured, and composable inside longer workflows. Extruct gives builders that layer across search, lookalikes, Deep Search, Deep Research, and AI Tables.

That fits Orthogonal well. We want agents to discover the right API, call it through one interface, and pay per request without setting up another vendor account. Extruct gives agents a company intelligence layer they can use for sales, sourcing, diligence, market mapping, and research workflows from day one.

Try It Today

Sign up for Orthogonal and get $10 free credits to try Extruct.ai and dozens of other APIs. One key, hundreds of APIs, pay per request.

Get Started | View on Orthogonal | Extruct.ai Website