How to Build an AI GTM Engineering Workflow (With the LeadOcean API)

AI GTM engineering means writing code and agents that run parts of outbound: finding accounts, enriching people, drafting messages, routing replies. The data layer comes first, and it is four API calls.

TL;DR

• A GTM engineer replaces manual list work with scripts and agents. The agent is only as good as the data it can query.
• The data layer is: size a list for free, search it, enrich the people you keep, then expose the same data to an agent over MCP.
• It takes about 30 minutes and spends about 40 records of the 1,000 on the free plan. No card.

Prerequisites

• A LeadOcean key from app.leadocean.io/register, stored as $LEADOCEAN_API_KEY.
• curl and jq.
• Optional: an MCP client such as Claude Code or n8n for step 4. See LeadOcean + n8n.

Requests go to https://api.leadocean.io with the header x-api-key. Free allows 5 requests a second.

Step 1: Size the list before you pull it

Sizing is free. Send count=true as a query parameter with limit=1. The body stays flat JSON and filters are arrays.

curl -s -X POST "https://api.leadocean.io/v1/people/search?count=true&limit=1" \
  -H "x-api-key: $LEADOCEAN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"title": ["GTM Engineer"], "country": ["US"]}' | jq '.meta'

meta.total is capped at 100,000. Data comes back empty and meta.countOnly is true. Nothing is spent.

Step 2: Search the people

Drop the count flag and set a small limit. One record is one person returned, so 25 people cost 25 records.

curl -s -X POST "https://api.leadocean.io/v1/people/search" \
  -H "x-api-key: $LEADOCEAN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"title": ["GTM Engineer"], "country": ["US"], "emailStatus": ["verified"], "limit": 25}' \
  > page1.json
jq '.meta' page1.json

Enum values are exact strings. Check them before you guess: GET /v1/enums, or the free leadocean_list_enum_values MCP tool. Page with cursor from meta. One search stops at 10,000 rows (meta.depthCapped), so narrow the filters instead of paging deeper.

Step 3: Enrich the people you keep

Search gives you a list. Enrich gives you one person in depth. It takes exactly one of linkedin_url, email, phone or person_id. Each search row carries its id at meta.source_ids.person_id.

PID=$(jq -r '.data[0].meta.source_ids.person_id' page1.json)
curl -s -X POST "https://api.leadocean.io/v1/people/enrich" \
  -H "x-api-key: $LEADOCEAN_API_KEY" \
  -H "Content-Type: application/json" \
  -d "{\"person_id\": \"$PID\", \"reveal_email\": true}"

Since October 2026 the response includes career history (resume_data): past roles, education and certifications. That is the raw material for a personalised first line. The enrich returns the person again, so it counts one record. A person we hold nothing for returns 404 and costs nothing.

Step 4: Give the same data to an agent over MCP

The hosted MCP server is at https://api.leadocean.io/mcp. It signs in with OAuth, so no key is pasted. It exposes 13 tools. Seven are free, including leadocean_count_leads, leadocean_count_companies, leadocean_list_filters and leadocean_list_enum_values.

In Claude Code, add it as a remote server (setup page):

claude mcp add --transport http leadocean https://api.leadocean.io/mcp

Then prompt the agent in plain language:

Use leadocean_count_leads to size US GTM Engineers with a mailable email.
Only if the total is under 500, search 10 of them and draft a two-line opener
from each person's career history. Do not export anything.

The guardrail in the prompt matters. Tell the agent which tools spend records and give it a ceiling.

Step 5: Move to a file when the list is big

An agent loop is the wrong tool above a few hundred rows. Use an export: POST /v1/exports takes up to 50,000 rows a request and 500,000 a day. Bulk CSV export is not on Free, so this step needs Solo ($97 a month) or higher.

curl -s -X POST "https://api.leadocean.io/v1/exports" \
  -H "x-api-key: $LEADOCEAN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"filters": {"title": ["GTM Engineer"], "country": ["US"], "emailStatus": ["verified"]}, "limit": 5000}'

The answer is a job id. Poll it, then download the CSV. Warm the sending domain before the first send: see inbox warm-up, Gmail warmup and inbox warming.

What you get

A real run of the free sizing tool on 21 October 2026 (leadocean_count_leads, title "GTM Engineer"). The counts include only mailable addresses (verified, catch_all_valid or catch_all), which is the tool's default.

Response shape from the API reference for an enrich, with placeholder values:

{
  "success": true,
  "data": {
    "profile_data": { "profile_full_name": "Jane Doe" },
    "contact_data": {
      "email_status": "verified",
      "contact_emails": [{ "email": "jane@acme.com", "status": "verified" }],
      "contact_current_experiences": [{ "job_title": "GTM Engineer", "company_name": "Acme", "company_domain": "acme.com" }]
    },
    "resume_data": {
      "experiences": [{ "job_title": "Sales Operations Analyst", "company_name": "Example Co" }],
      "educations": [],
      "certifications": []
    }
  },
  "meta": { "credits": 1 }
}

Past employers in resume_data carry no company_id yet.

Troubleshooting

FAQ