Jaspor API

Welcome to the Jaspor API. your direct access to Jaspor’s dynamic dataset and intent-driven networking engine on X.com.

This API is designed for builders, founders, and developers who want to plug Jaspor’s intelligence and network into their own products, dashboards, and workflows. You bring the idea, we provide the signal, intros, and proof.

All endpoints are JSON over HTTPS, versioned at /v1/. Authentication is via API keys (bearer tokens).

Technical Overview

The Jaspor API is powered by a multi-stage AI Matchmaker Algorithm:

Input Data Ingestion
- User profiles: structured metadata (roles, geos, preferences).
- Behavioral signals: interaction history (likes, replies, dwell time).
- Semantic embeddings: public text inputs encoded with multi-model LLMs (SBERT, LLaMA-derived).
Feature Normalization & Weighting
- Normalize features into a shared latent space.
- Apply hard constraints (geo, role, filters) and soft constraints (interests, vibe scoring).
- Reinforcement updates from positive matches and sustained conversations.
Candidate Retrieval
- Approximate nearest neighbor (ANN) search with FAISS/HNSW.
- Hybrid scoring function:
  $S(u,v) = \alpha \cdot \cos(E_u, E_v) + \beta \cdot f_{\text{compat}}(u,v) - \lambda \cdot \text{penalties}$
Ranking & Explainability
- Cross-encoder re-ranking or lightweight neural ranker on top-K.
- Gradient-based attribution highlights “why” behind each match.
Multi-Model LLM Stack
- Small LLM: parses intent and entities.
- Mid-sized LLM: drafts intros and smart DMs.
- Policy LLM: ensures safety, tone, and disclosure.

Top 10 Endpoints

These 10 endpoints showcase Jaspor’s unique value: intent-based matching, real-time discovery, warm intros, and verifiable social graph data.

1. Create an Intent

POST /v1/intents

Define who or what you’re looking for. Jaspor interprets your goal and starts building matches instantly.

Request

{
  "goal": "Find AI founders building in design tools",
  "tags": ["ai", "design", "founders"],
  "geos": ["NA", "EU"],
  "timeframe": "next_30_days",
  "notes": "Prefer early-stage, no agencies"
}

Response

{
  "intent_id": "int_12345",
  "status": "active",
  "created_at": "2025-09-28T12:00:00Z"
}

2. Get Recommended Matches

POST /v1/match/recommendations

Retrieve ranked candidates Jaspor thinks you should connect with.

Request

{
  "intent_id": "int_12345",
  "limit": 10,
  "diversity": "topic"
}

Response

{
  "intent_id": "int_12345",
  "results": [
    {
      "candidate_id": "x_7781",
      "handle": "@janedoe",
      "score": 0.94,
      "why": ["Seed-stage founder", "Active on AI/UX threads", "Mutuals: @sarahdev"]
    },
    {
      "candidate_id": "x_9912",
      "handle": "@founderAI",
      "score": 0.88,
      "why": ["Recently launched AI design plugin"]
    }
  ]
}

3. Explain a Match

GET /v1/matches/:id

Fetch Jaspor’s “why” behind a particular match.

Response

{
  "match_id": "match_abc789",
  "candidate": "@janedoe",
  "reasoning": [
    "Active AI founder in design tools",
    "Mutual followers with @you",
    "Engaged with similar content in the past 7 days"
  ],
  "score": 0.94
}

4. Create a Warm Intro

POST /v1/intros

Spin up a Jaspor-crafted group DM with the perfect opener.

Request

{
  "intent_id": "int_12345",
  "candidate_ids": ["x_7781","x_9912"],
  "platform": "x",
  "opener_style": "friendly",
  "cc_requester": true
}

Response

{
  "intro_id": "intro_4567",
  "status": "draft",
  "suggested_opener": "Hey all — given your shared focus on AI/UX design tools, thought it made sense to connect you."
}

5. Send an Intro

POST /v1/intros/:id/send

Finalize and send the group DM on X.

Response

{
  "intro_id": "intro_4567",
  "status": "sent",
  "message_url": "https://x.com/messages/1234"
}

6. Smart DM Draft

POST /v1/prompts/message

Generate a context-aware opener for a candidate based on their latest X activity.

Request

{
  "candidate_id": "x_7781",
  "tone": "concise",
  "context": "They just shipped an AI Figma plugin",
  "goal": "Explore collab opportunities"
}

Response

{
  "draft": "Congrats on the Figma plugin launch 🚀 Curious how you're thinking about design-to-compliance workflows."
}

7. Auto-Circle Creation

POST /v1/circles

Create a curated “mastermind” group from intent + network data.

Request

{
  "name": "AI Founders – NYC",
  "seed_intent_id": "int_12345",
  "membership": { "size": 8, "diversity": "geo" },
  "platform": "telegram"
}

Response

{
  "circle_id": "circle_9099",
  "members": ["@janedoe","@founderAI","@sarahdev"],
  "kickoff_message": "Welcome! This group is curated around AI founders in NYC."
}

8. On-Chain Proof of Intro

POST /v1/onchain/intros/:id/proof

Generate a Solana transaction proving an intro occurred.

Response

{
  "intro_id": "intro_4567",
  "tx_base64": "AAAA...",
  "required_signers": ["7XhP9...abc"],
  "expires_at": "2025-10-01T00:00:00Z"
}

9. Reputation & Badges

GET /v1/onchain/reputation/:wallet

Fetch Jaspor’s verifiable reputation graph (NFT badges, scores).

Response

{
  "wallet": "7XhP9...abc",
  "badges": [
    { "type": "SuperConnector", "minted_at": "2025-09-20T00:00:00Z" },
    { "type": "Top AI Founder", "minted_at": "2025-09-25T00:00:00Z" }
  ],
  "score": 87
}

10. Real-Time Opportunity Radar

GET /v1/radar/opportunities

Surface trending conversations and people on X that align with your intent.

Response

{
  "intent_id": "int_12345",
  "opportunities": [
    {
      "thread_url": "https://x.com/someone/status/123",
      "summary": "Discussion on AI design tools blowing up",
      "recommended_action": "Engage with @janedoe and @uxAI"
    },
    {
      "thread_url": "https://x.com/other/status/456",
      "summary": "Seed investors sharing interest in AI productivity apps"
    }
  ]
}

Usage Notes

Rate limits: 100 requests/minute per key.
Auth: Authorization: Bearer <api_key>.
Idempotency: Use Idempotency-Key headers for POST requests.
Errors: Standardized JSON:

{ "code": "invalid_request", "message": "Intent not found" }

PreviousJaspor AI NextVision

Last updated 4 days ago