// products
People Search APIPeople Enrichment APICompany Data APICompany Enrichment APILLM Data APISales MCP Server
// solutions
AI SDR APISales AutomationCandidate SourcingVC Data APILead EnrichmentBuying Signals
// explore
FeaturesIndustriesPricingCompliance
Sign inStart for free
// fig. 01 — agentic search

B2B Data API
for LLM agents.

Send a sentence. The endpoint plans the search, picks the filters, runs it, and returns verified B2B data ranked by intent — built for AI agents.

Try the agent for freeSee the API →
↓ scaleplanner + searcher + reranker
1 call// any query
auto// filter plan
142ms// p50 first token
API
// figure-02 — endpoint

One sentence in. A verified result set out.

The agent translates natural language into filter plans, decomposes multi-step queries, and reranks the output by intent — all in a single request.

Send a sentence. We translate it into structured filters, run the search, and return ranked results — explanation included.

POST/v1/agent/search
// 03 · request
{
  "query": "VPs of Sales at Series B SaaS in NYC, last 18 months",
  "max_results": 10,
  "agent": "claude-3-5-sonnet"
}
200200 OK · application/json
// 04 · response
{
  "plan": {
    "filters": {
      "current_title": { "type": "like", "value": "VP Sales" },
      "current_company_funding_stage": { "type": "=", "value": "series_b" },
      "current_company_industry": { "type": "=", "value": "Software" },
      "current_job_location": { "type": "like", "value": "New York" },
      "years_in_current_position": { "type": "<=", "value": 1.5 }
    }
  },
  "results": [
    {
      "id": "prof_PVURNGJihYludzaOwFC",
      "name": "Priya Anand",
      "title": "VP Sales · Ramp",
      "intent_score": 0.94,
      "explanation": "Joined 11 months ago; Ramp closed Series B in Q3."
    }
  ],
  "credits_used": 3,
  "tokens": { "in": 412, "out": 1840 },
  "cost_usd": 0.0021
}
Get API accessBrowse capabilities →
Capabilities
// capabilities

An agent that plans, searches, ranks.

It reads the query, picks the tool, builds the filter payload, runs it, and ranks the results against the original intent — every step traceable.

// fig. 05─ ─ ─

Plan, then search

The agent reads the query, picks the right tool (people / company / enrich), builds the filter payload, then runs it.

// filters
search_peoplesearch_companiesenrich_profile
// fig. 06─ ─ ─

Multi-step search

Compose searches: find the companies, then the people inside, then enrich. The agent chains tool calls without your code.

// filters
stepstool_chaincontext_pass-through
// fig. 07─ ─ ─

Intent reranking

Every result returns a 0–1 intent score against the original natural-language query — sort or threshold on this, not on order.

// filters
intent_scoreexplanationsignals
// fig. 08─ ─ ─

Cost & traceability

Every response ships token counts, credit usage and the exact filter plan — auditable, reproducible, billable per run.

// filters
plancredits_usedtokenscost_usd
Coverage
// fig. 10 — coverage

Source data across 20+ points

Continuous global refresh from public sources — not a stale quarterly snapshot. Profiles, companies, and contact data verified in real time.

// dataset · global feed20+ points
// sources
20+
// refresh
live
// regions
global
Delivery
// fig. 11 — delivery

Data delivered where your stack lives.

Same dataset, three transports. Pull it via REST, plug it into your agent through MCP, or stream changes to your webhook in real time.

// fig. 11REST
// channel

API

Pull on demand.

One JSON request. 60+ columns. 8 operators. Authenticate with a Bearer token, point at the endpoint, get verified data back in 142ms (p50).

POST/v1/people/search
// notes
  • Bearer auth
  • JSON in, JSON out
// fig. 12AGENT
// channel

MCP

Native tools for agents.

Drop our MCP server into Claude, Cursor, or any agent runtime. The model picks the tool, builds the filters and reads the response — no glue code, no schemas to maintain.

TOOLsearch_people · enrich_profile
// notes
  • Claude · OpenAI · Cursor
  • tool-call native
  • no schema babysitting
// fig. 13PUSH
// channel

Webhooks

Get notified the moment it changes.

Subscribe to ICP events — new matches, role changes, funding rounds. We POST to your endpoint with the delta. Your sequence reacts before the data goes stale.

POST→ your_endpoint
// notes
  • ICP delta events
  • signed payloads
  • exponential retry
Use Cases
// fig. 14 — build surface

What can you build with our API.

One POST request, four product surfaces. Each layered on top of the same endpoint — only the question changes.

// fig. 15sourcing copilots

AI Recruiting product

Past employers, top schools, degrees, certifications and tenure in the same payload. Your agent shortlists ex-FAANG, ex-Big4 and Ivy/MIT graduates — and the data refreshes itself.

// filters
past_companyschooldegree_levelskillcertificationyears_of_experience
// outputranked shortlist → ATS
// fig. 16outbound agents

AI SDR

Wire funding triggers and ICP filters into your agent. Stream verified profiles into the sequence. Re-query when the ICP changes — the agent owns the list, not a CSV exported in March.

// filters
current_company_funding_stagecurrent_titlecurrent_company_sizeprofile_countryis_currently_employed
// outputlive prospect feed → sequence
// fig. 17deal sourcing

Investment platform

Catch stealth founders, track investor overlap, funding stage and tenure across thousands of operators. Pipe raw JSON into your scoring model — no scrapers, no stale lists, no broken refresh jobs.

// filters
current_title ~ "Stealth Founder"current_company_investorcurrent_company_funding_stagecurrent_company_sizeyears_at_current_companycurrent_company_industry
// outputscored deal flow → analyst stack
// fig. 18embedded data layer

Sales automation platform

Power people search, enrichment and ICP refresh inside your product — one endpoint, MCP-ready, GDPR-compliant. Stop maintaining brittle scrapers; ship the feature instead.

// filters
current_companycurrent_titlecurrent_company_categorylanguage_proficiencyfollower_count
// outputembedded search → your app

Building something we haven't listed? Tell us what you're wiring it into. Half the columns we ship today came from a customer asking for one.

Compliance
GDPR Logo

GDPR compliant

Full compliance with European data protection regulations. Your data privacy is our priority.

CCPA Logo

CCPA compliant

Adhering to California Consumer Privacy Act standards for maximum data protection.

Publicly available data

All data sourced from publicly available and verified sources, ensuring ethical collection.

Related
// see also

Related solutions

fig. A

AI SDR API

Outbound agents use natural-language search to source the next prospect.

Explore
fig. B

Sales MCP Server

Same dataset exposed as MCP tools for Claude, Cursor and Cline.

Explore
fig. C

Embedded B2B Data API

Embed natural-language search inside your sales platform UI.

Explore
Get Started
// fig. ∞ — ship

Build with us. Now.

Get an API key in 60 seconds. Plug your AI agent into 800M+ verified profiles and 75M+ companies — today.

Get an API keyTalk to founders
↓ nextREST · MCP · Webhooks