Your platform deserves real-time financial intelligence. TerminalTrend processes earnings calls, SEBI filings, and market news — returning structured JSON in under 500ms via a single API call.
Earnings calls, regulatory filings, and market news hold critical signals — but extracting them reliably at the speed markets demand remains beyond reach for most platforms.
Structured financial data APIs from institutional providers carry price points that exclude mid-market fintech companies and independent brokers.
Analyst-reviewed transcripts arrive hours after events conclude — a competitive liability for trading and advisory platforms.
Models trained on broad web text frequently misinterpret accounting standards, regulatory terminology, and forward guidance language.
Building in-house NLP for financial documents requires specialised talent, months of development, and ongoing model maintenance.
The following criteria separate capable vendors from overpromised ones. TerminalTrend meets all of them.
Early integration partners get dedicated onboarding, preferred API pricing, and direct engineering support. Limited slots available — we are onboarding 10 platforms in this cohort.
terminaltrend.io · founder@terminaltrend.io · India
TerminalTrend sits between raw financial documents and your product — processing, structuring, and delivering intelligence at the speed your users expect.
Every financial document — earnings call transcript, regulatory filing, or news article — passes through a three-stage pipeline: ingestion, NLP inference, and structured output delivery.
The ingestion layer accepts raw text or document URLs. Inference runs on dedicated NVIDIA GPU instances using models fine-tuned on domain-specific financial corpora. Output is a consistent JSON schema regardless of input document type.
Unlike general-purpose AI APIs that use shared compute and unpredictable response times, TerminalTrend runs on reserved GPU infrastructure. Each API request is handled by a dedicated inference worker with a p99 latency guarantee of under 800ms for documents up to 150 pages.
The pipeline uses NVIDIA NeMo for fine-tuning and TensorRT for optimised inference — the same GPU stack used by institutional trading firms, made accessible via a standard REST API.
Indian market coverage is a deliberate design priority, not an afterthought. The NLP models are fine-tuned on SEBI disclosures, BSE/NSE filings, DRHP documents, and Indian earnings call transcripts, in addition to global SEC filings and international financial news.
Client documents are processed entirely in-memory and are not stored, logged, or used for any form of model retraining following inference. All API communications are HTTPS-only with TLS 1.3. We operate on SOC2-aligned infrastructure, and a Data Processing Agreement is available for enterprise contracts on request.
Each module is independently fine-tuned on financial-domain data. Deploy one or all — usage scales with your platform.
| Document Type | Source | Outputs |
|---|---|---|
| Earnings call transcript | BSE, NSE, SEC EDGAR, direct upload | Sentiment, guidance, metrics, tone, risk flags |
| Annual report / 10-K | SEC EDGAR, SEBI, company website | Financial highlights, risk language, segment data |
| Quarterly filing / 10-Q | SEC EDGAR, SEBI disclosures | QoQ changes, material events, guidance updates |
| 8-K / material event | SEC EDGAR, BSE/NSE announcements | Event classification, impact assessment, entities |
| DRHP / IPO prospectus | SEBI, stock exchange filings | Risk factors, financials, promoter data, use of funds |
| Financial news article | Real-time feed or URL | Entity sentiment, topic classification, key facts |
A single REST endpoint accepts earnings transcripts, regulatory filings, or news articles and returns structured intelligence with a consistent response format.
https://api.terminaltrend.io/v1
All requests require a Bearer token in the Authorization header. API keys are issued on account creation and can be rotated from the dashboard at any time.
Authorization: Bearer tt_live_xxxxxxxxxxxxxxxxxxxxxxxx
Analyze any financial document. Returns structured JSON intelligence in under 500ms on standard documents.
{
"type": "earnings_call", // earnings_call | filing | news
"ticker": "INFY", // optional: NSE/BSE/NYSE/NASDAQ symbol
"text": "<document content>", // raw text or transcript
"market": "IN" // IN | US | GB | SG — default: auto
}
{
"id": "inf_01jx8k2p4n...",
"type": "earnings_call",
"ticker": "INFY",
"sentiment": "positive",
"sentiment_score": 0.82,
"guidance": {
"revenue_growth": "8-10% YoY",
"margin_outlook": "stable",
"next_quarter_eps": "18.40-18.80"
},
"risk_flags": ["FX headwinds", "client concentration"],
"key_metrics": {
"eps": 18.42,
"revenue_growth_yoy": 9.1,
"ebitda_margin": 24.1,
"headcount_change": "flat"
},
"management_tone": "confident",
"latency_ms": 340,
"model_version": "tt-fin-nlp-v2.1"
}
| Parameter | Type | Required | Description |
|---|---|---|---|
| type | string | Yes | Document type: earnings_call, filing, or news |
| text | string | Yes | Raw document text. Maximum 500,000 characters (~150 pages). |
| ticker | string | No | Exchange ticker symbol. Improves entity resolution accuracy. |
| market | string | No | Market context: IN, US, GB, SG. Auto-detected from content if omitted. |
| extract | array | No | Specific outputs to include. Defaults to all. Options: sentiment, guidance, metrics, risk_flags, tone. |
| Plan | Requests/min | Max document size | P95 latency SLA |
|---|---|---|---|
| Starter | 30 req/min | 50 pages | Best effort |
| Scale | 200 req/min | 150 pages | <800ms |
| Enterprise | Custom | Unlimited | <500ms dedicated |
TerminalTrend was founded to bring institutional-grade financial intelligence within reach of mid-market fintech companies and independent brokers across India and global emerging markets.
Structured financial intelligence has historically been the domain of large institutional players with the resources to build proprietary data operations or pay for enterprise data terminal access. TerminalTrend exists to change that.
We believe that mid-market fintech platforms, independent brokers, and emerging-market financial products deserve the same quality of real-time, structured financial intelligence that institutional desks take for granted — delivered as a simple API rather than a multi-year procurement and integration project.
India represents one of the most dynamic and underserved markets for financial intelligence infrastructure. The volume of SEBI filings, BSE/NSE disclosures, and Indian earnings call transcripts is substantial — yet most AI financial intelligence vendors treat Indian market coverage as an afterthought.
Our NLP models are trained on Indian financial documents from the ground up, with native understanding of SEBI regulatory language, Ind-AS accounting standards, and the specific disclosure patterns of Indian listed companies.
TerminalTrend is built on NVIDIA NeMo for model fine-tuning and TensorRT for production inference, deployed on dedicated AWS EC2 GPU instances in the ap-south-1 (Mumbai) region for India-market customers. The inference pipeline is purpose-built for financial document processing — not a general-purpose LLM wrapper.
Perspectives on NLP for financial markets, API design for broker platforms, and the state of AI in emerging market fintech.
We are a small, technical team building AI infrastructure for fintech. We hire people who care deeply about craft and impact.
We are a remote-first team based primarily in India, with deep roots in the Chennai fintech and AI research community. We move fast, ship real products, and take the quality of our models seriously.
Whether you're evaluating the API, requesting a live demonstration, or ready to start integration — reach out and we'll respond within one business day.
See TerminalTrend process your own financial documents live. We'll walk through the API, discuss your integration requirements, and answer technical questions.
founder@terminaltrend.ioQuestions about the API, latency requirements, data handling, or custom enterprise integration needs. Our engineering team responds directly.
founder@terminaltrend.ioWe respond to all enquiries within one business day. Your information is not shared with third parties.
Last updated: June 2025
TerminalTrend collects information you provide directly — including your name, email address, company name, and the financial documents you submit for analysis via our API. We also collect standard server logs including IP addresses, request timestamps, and API response codes for security and operational purposes.
We use contact information to communicate with you about your account, respond to enquiries, and send service-related updates. Financial documents submitted to the API are processed in-memory for inference and are not stored, retained, or used for model training following the completion of each API request.
API request logs are retained for 30 days for security and debugging purposes. Contact information is retained for the duration of your relationship with TerminalTrend and deleted upon request. Document content submitted for inference is not retained after the response is returned.
We do not sell, rent, or share your personal information with third parties for marketing purposes. We may share data with infrastructure providers (AWS) strictly for the purpose of operating the service. Enterprise customers may request a Data Processing Agreement (DPA) at any time.
All API communications are encrypted using TLS 1.3. We operate on SOC2-aligned infrastructure with access controls, audit logging, and regular security reviews. Document content is never written to persistent storage during inference.
For privacy enquiries or data deletion requests, contact us at founder@terminaltrend.io.
Last updated: June 2025
By accessing or using the TerminalTrend API and related services, you agree to be bound by these Terms of Service. If you are using the service on behalf of an organisation, you represent that you have the authority to bind that organisation to these terms.
You may use the TerminalTrend API to analyse financial documents for lawful business purposes within the rate limits and document size constraints of your subscription plan. You may not use the service to process documents you do not have the right to analyse, or to circumvent any applicable financial data regulations in your jurisdiction.
Your use of the API is subject to the rate limits and document size constraints of your subscription plan. Exceeding rate limits will result in HTTP 429 responses. Sustained abuse of rate limits may result in account suspension.
The structured JSON output returned by the API for documents you submit is yours to use in your products and services. The underlying NLP models, inference infrastructure, and API design remain the intellectual property of TerminalTrend Technologies.
TerminalTrend provides AI-generated financial intelligence for informational purposes. The output should not be construed as financial advice. TerminalTrend is not liable for investment or trading decisions made on the basis of API output.
For terms enquiries, contact founder@terminaltrend.io.
TerminalTrend is built on enterprise-grade security practices appropriate for financial data processing workloads.
All API communications use TLS 1.3. API keys are transmitted only in request headers, never in URLs. Webhook payloads are signed with HMAC-SHA256.
Financial documents submitted for inference are processed entirely in-memory and are not written to any persistent storage. No document content is retained following the API response.
We operate on AWS infrastructure with access controls, audit logging, and network segmentation consistent with SOC2 Type II principles. Enterprise customers may request our security documentation.
API keys can be rotated at any time from the dashboard. Keys are stored as salted hashes — we cannot recover a key once generated. Compromised keys can be revoked instantly.
India-market customers are served from the AWS ap-south-1 (Mumbai) region by default. Data does not leave the India region during inference unless explicitly requested.
Enterprise customers may request a full DPA for their compliance requirements.
When you submit financial documents via the TerminalTrend API, TerminalTrend acts as a data processor on your behalf. You remain the data controller responsible for the lawful basis of processing any personal data contained in the documents submitted.
TerminalTrend processes submitted document text solely for the purpose of generating structured NLP inference output as described in the API Reference. No processing activity beyond inference execution is performed on document content.
TerminalTrend uses Amazon Web Services (AWS) as its primary infrastructure provider. Document content may pass through AWS compute and memory resources in the ap-south-1 (Mumbai) region during inference. AWS acts as a sub-processor under appropriate contractual protections.
Enterprise customers requiring a formal Data Processing Agreement for regulatory compliance (GDPR, DPDP Act India, or equivalent) should contact founder@terminaltrend.io. We aim to return a signed DPA within 5 business days.