Describe the people you need in plain English. Get matching profiles with LinkedIn and optional emails and phone numbers, plus matching X, Instagram, TikTok and Facebook accounts and YouTube appearances, checked against their professional profile.
1,000 free credits · No credit card
{
"people": [
{
"name": "Alex Morgan",
"title": "VP of Sales",
"company": "Solstice Energy",
"headline": "VP of Sales at Solstice Energy",
"location": "San Francisco, CA",
"linkedin_url": "https://www.linkedin.com/in/alex-morganexample"
},
{
"name": "Priya Raman",
"title": "Head of Sales",
"company": "Helio Grid",
"headline": "Head of Sales at Helio Grid",
"location": "Oakland, CA",
"linkedin_url": "https://www.linkedin.com/in/priya-ramanexample"
}
// …2 more
],
"ms": 11420
}
{
"people": [
{
"name": "Ciara Byrne",
"title": "Engineering Manager",
"company": "Northwind Labs",
"headline": "Engineering Manager at Northwind Labs",
"location": "Dublin, Ireland",
"linkedin_url": "https://www.linkedin.com/in/ciara-byrneexample"
},
{
"name": "Tomás Kelly",
"title": "Senior Engineering Manager",
"company": "Northwind Labs",
"headline": "Senior Engineering Manager at Northwind Labs",
"location": "Dublin, Ireland",
"linkedin_url": "https://www.linkedin.com/in/tom-s-kellyexample"
}
// …2 more
],
"ms": 14870
}
{
"people": [
{
"name": "Lena Weber",
"title": "Co-founder & CEO",
"company": "Carbonloop",
"headline": "Co-founder & CEO at Carbonloop",
"location": "Berlin, Germany",
"linkedin_url": "https://www.linkedin.com/in/lena-weberexample"
},
{
"name": "Jonas Richter",
"title": "Founder",
"company": "Gridwise",
"headline": "Founder at Gridwise",
"location": "Berlin, Germany",
"linkedin_url": "https://www.linkedin.com/in/jonas-richterexample"
}
// …2 more
],
"ms": 9630
}
One query, every place a person shows up
LinkedIn profiles
X
Instagram
TikTok
Facebook
YouTube videosSame API key, same credits. Search turns a description into a shortlist; Enrich turns the person you meant into a full, verified record.
A sentence in, as many people as you ask for out, best match first. Each with name, title, employer, headline, location, and LinkedIn address.
A LinkedIn address in: work history, education, optional emails and phones, and the social accounts and videos that are really them.
@alexmorganX · 4.1K followersmatch@alex.morgan.sfInstagram · 1.2K followersmatchSelling solar to utilitiesYouTube · panel, 18 minmatch@alexmorgan.salesTikTok · 310 followersmatchWrite it the way you’d brief a recruiter. An agent plans the filters, looks up each company, and loosens the search if nothing fits.
Each account and video is compared with the person’s background. Only confident matches come back, not everyone with the same name.
Turn on contact lookup to include work and personal emails with deliverability, and phone numbers. Off by default; bounced addresses are left out.
Your description is matched against professional profiles. Pick the one you meant from up to 20.
Enrich pulls the full profile: headline, location, work history, and education. Turn on contact lookup to add emails and phones.
X, Instagram, TikTok, Facebook and YouTube are searched in parallel by name. YouTube also searches the name with their employer.
A model reads every account and video against the profile, and only the ones that are them come back.
Two POST requests with a Bearer key. Or connect the MCP server and Claude Code, Cursor, or any MCP client gets search_people and enrich_person as tools, billed like the API.
https://www.constants.io/people-search/mcp# Find people from a description curl -X POST "https://www.constants.io/people-search/v1/people/search" \ -H "Authorization: Bearer $CONSTANTS_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "query": "VPs of sales at solar companies in California", "max_results": 10 }' # Enrich the one you meant curl -X POST "https://www.constants.io/people-search/v1/people/enrich" \ -H "Authorization: Bearer $CONSTANTS_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "linkedin_url": "https://www.linkedin.com/in/alex-morgan-example" }'
const call = (path, body) => fetch("https://www.constants.io/people-search" + path, { method: "POST", headers: { Authorization: `Bearer ${process.env.CONSTANTS_API_KEY}`, "Content-Type": "application/json" }, body: JSON.stringify(body), }).then((r) => r.json()); const { people } = await call("/v1/people/search", { query: "VPs of sales at solar companies in California", }); const { person, platforms, videos } = await call("/v1/people/enrich", { linkedin_url: people[0].linkedin_url, }); console.log(person.name, platforms.x.accounts[0]);
import os, requests def call(path, body): r = requests.post("https://www.constants.io/people-search" + path, json=body, headers={"Authorization": f"Bearer {os.environ['CONSTANTS_API_KEY']}"}) r.raise_for_status() return r.json() people = call("/v1/people/search", {"query": "VPs of sales at solar companies in California"})["people"] profile = call("/v1/people/enrich", {"linkedin_url": people[0]["linkedin_url"]}) print(profile["person"]["name"])
# Claude Code: search_people and enrich_person as tools claude mcp add --scope user --transport http constants-people-search \ https://www.constants.io/people-search/mcp \ --header "Authorization: Bearer $CONSTANTS_API_KEY" # Cursor and other clients: ~/.cursor/mcp.json { "mcpServers": { "constants-people-search": { "url": "https://www.constants.io/people-search/mcp", "headers": { "Authorization": "Bearer cps_…" } } } }
Turn your ideal customer profile into decision makers with verified work emails, ready for your outbound tool.
Find people by past employer, school, and city, then see what they post and talk about before you reach out.
Find the videos someone actually speaks in, their real accounts, and what they’re about, for guests, speakers, or partners.
Every call spends credits for what it costs to run, about 14 for each person a search finds. Credits are shared with your workspace's simulations.
Try every endpoint and the MCP server.
For founders and small teams.
For product teams shipping weekly.
For custom volume, security and compliance.
Need more? Credit packs: $20 for 2,000 · $50 for 5,000 · $100 for 10,000 · $250 for 25,000 credits. They last 12 months.
It finds people who match a description and returns them as structured data. Send a sentence like “Engineering managers at Northwind in Dublin” and get back as many people as it asks for (20 by default, up to 5,000), each with their name, current title, employer, location, and LinkedIn address.
An agent reads your description, looks up each company it names, or finds companies by funding, revenue, size, or industry when you describe them instead, and runs filtered searches by name, current or past employer, title, location, school, or seniority, using the exact values the database stores. It checks small samples, loosening or tightening the filters until they fit, then pages through the matches for as many people as you asked for.
Give it a LinkedIn address for the professional profile, matched X, Instagram, TikTok and Facebook accounts, and YouTube videos. Emails and phone numbers are optional and off by default. Give it a Reddit profile URL for the public profile, avatar, recent posts and comment previews, with more activity available through pagination.
A model compares every candidate account and video with the person’s professional background. Only the ones it’s confident are the person come back, so you get an answer, not a list of everyone with the same name.
Each call spends credits worth about 1.43× what it cost to run in data and model time, at $0.01 a credit, at least one a call: about 7 credits for each person a search finds. Your usage page shows every call and its credits. A search usually takes 8–20 seconds, longer when it searches companies too.
Yes. The service is an MCP server with two tools, search_people and enrich_person. Add it to Claude Code, Cursor, or any MCP client with your API key; tool calls are billed exactly like API calls.
Yes. Every account gets 1,000 free credits, with no credit card. Choose a plan, or buy a credit pack, when you need more.
Every month. No credit card required.