Documentation: Expert Network Transcripts API

Overview

The Nextmark Expert API provides access to an exclusive library of 2,500+ primary research interviews generated by our internal Expert Network (EN) and Research & Consulting (R&C) divisions. Unlike public earnings calls, these are deep-dive due diligence interviews with industry insiders.

Key Capabilities:

Source: Proprietary internal research (Exclusive).
Enrichment: NLP-tagged for Company Mentions, Products, and Context.
Compliance: Rigorously scrubbed for MNPI and PII.
Granularity: Speaker profile metadata (Role, Tenure, Competitor History).

‍

1. Access Methods

Method	Use Case	Format
REST API	Searching for specific insights ("Databricks vs Snowflake").	JSON
Bulk Feed	Ingesting full history for proprietary LLM training.	JSONL
Vector Feed	New: RAG. Semantic search across the library.	Embeddings

‍

2. REST API: Expert Transcripts Endpoint

Endpoint: GET https://api.nextmark.data/v1/expert-transcripts

Request Parameters:

Parameter	Type	Required	Description
api_token	string	Yes	Your API Key.
ticker	string	No	Filter by primary subject (e.g., SNOW).
mentioned_ticker	string	No	Find transcripts where a company was mentioned, even if not the main topic.
expert_role	string	No	Filter by role: Former_Exec, Customer, Partner.
topic_tag	string	No	Filter by context: Pricing, Competition, Churn.

‍

Example Request (Python)

import requests

url = "https://api.nextmark.data/v1/expert-transcripts"
params = {
    "api_token": "YOUR_KEY",
    "ticker": "SNOW",
    "expert_role": "Former_Exec"
}

response = requests.get(url, params=params)
data = response.json()

‍

Request Structure (JSON)

{
  "transcript_id": "EXP-4921",
  "source_division": "Nextmark_Consulting_Group",
  "primary_ticker": "SNOW",
  "interview_date": "2025-01-15",
  "expert_profile": {
    "role": "Former VP Sales",
    "tenure": "2020-2024",
    "verification_status": "Verified"
  },
  "mentions_summary": [
    {"ticker": "SNOW", "sentiment": -0.4, "count": 12},
    {"ticker": "AMZN", "sentiment": 0.2, "count": 4}
  ],
  "segments": [
    {
      "speaker": "Expert",
      "text": "Databricks is definitely winning more Proof of Concepts (POCs) in the enterprise layer.",
      "nlp_tags": {
        "entities": ["Databricks", "Enterprise"],
        "context": "Competitive_Loss",
        "sentiment_score": -0.85
      }
    }
  ]
}

‍

3. Data Dictionary (Key Metrics)

Field	Type	Description
source_division	String	Origin of the interview. Nextmark_EN (Expert Network) or Nextmark_RC (Research & Consulting).
mentions_summary	List	Alpha Signal. Aggregated sentiment for every company mentioned in the call. Allows for "Cross-Impact" analysis (e.g. Bullish for Competitor = Bearish for Target).
nlp_tags.context	String	Proprietary categorization of the specific paragraph. Tags include: Pricing_Power, Supply_Chain, Management_Quality.
verification_status	String	Confirmation that the expert's employment history was validated by our compliance team.

‍

4. NLP Enrichment & Context Highlighting

Because these transcripts originate from our internal divisions, we apply advanced metadata tagging before release.

Feature	Function	Use Case
Entity Extraction	Identifies Tickers, Private Companies, and Products.	"Find all calls mentioning 'Snowpark' product."
Contextual Sentiment	Calculates sentiment per entity in the sentence.	"Expert loves the Tech (Positive) but hates the Price (Negative)."
Competitor Mapping	Links the expert's prior employers to current insights.	"Former AWS employee discussing Google Cloud architecture."

‍

5. Bulk Data Access (JSONL)

Download the full library for offline analysis.

File Naming Convention: Nextmark_Expert_Transcripts_{Sector}_{Year}.jsonl

Format: JSON Lines. Each line is a full interview object with nested segments.

How to Download:

Navigate to Data Export.
Select Expert Research Package.
Choose Sector (e.g. "SaaS", "Healthcare").
Click Download ZIP.

‍

6. Vector Database Feed (RAG-Ready)

Exclusive to Nextmark. This feed enables "Question Answering" across the expert library using semantic search.

Sample Vector Output (Decoded):

"Context: Former Sales VP at Crowdstrike (Interview Jan 2025). Topic: Win Rates. Text: 'We saw Microsoft Defender win rates increase in the SMB segment due to bundling.' Tags: [Competition, Bundling, SMB]"

Use Case:

"Summarize the top 3 complaints customers have about Salesforce."
"What are former employees saying about the new CFO's cost-cutting measures?"

‍

7. Error Handling

401 Unauthorized: Invalid API Key or Subscription Tier (Expert Data is Tier 3+).
403 Forbidden: Access restricted due to conflict of interest (configurable for Enterprise clients).
429 Too Many Requests: Rate limit exceeded.

‍

8. Need Help?

Research Desk: email research-support@nextmark.ai
Developer Support: email dev-support@nextmark.ai

Author Name

Team Nextmark