EarningsNerd
AI-powered SEC filing analysis. EarningsNerd turns dense 10-K and 10-Q filings into clear,
evidence-backed summaries — business performance, financials, risks, and management discussion —
so investors can understand a company in minutes instead of hours.
🌐 Live: earningsnerd.io · API: api.earningsnerd.io
🔒 Status: pre-launch (waitlist mode)
What it does
- Search any public company by name or ticker and pull its filings straight from SEC EDGAR.
- Generate structured AI summaries of 10-Ks and 10-Qs, streamed in real time — executive
snapshot, financials, liquidity, MD&A, risks, guidance, and trends.
- Evidence-backed output. Every bullet must cite a filing excerpt or an XBRL anchor; the AI
output is held to a strict JSON contract with deterministic repair and a fallback path.
- Side-by-side comparison of 2–5 filings to surface trends and changes (Pro).
- Ask this Filing (Pro) — natural-language Q&A grounded in a single filing, with deep-linked
citations back to the exact passage or XBRL figure.
- Financial visualization — interactive charts for revenue, earnings, and key metrics, grounded
in parsed XBRL data.
- Insider activity & peer benchmarking — Form 4 open-market trades and cross-company metric
comparisons, drawn from parsed XBRL.
- Personalized dashboard — a watchlist-driven feed of what changed across your companies, plus
an upcoming-earnings calendar.
- Watchlists, saved summaries, trending & hot filings, and email alerts.
- Tiered subscriptions (Free / Pro) via Stripe.
Tech stack
Frontend — Next.js 16 (App Router) · TypeScript · Tailwind CSS + shadcn/ui · React Query ·
Recharts · PostHog · Sentry. Deployed on Vercel.
Backend — FastAPI (Python 3.11) · SQLAlchemy 2.0 · PostgreSQL 15 · two-tier cache (in-memory +
optional Redis). SEC data via edgartools (XBRL extraction, filing parsing). Deployed on
Google Cloud Run with Cloud SQL.
AI — an OpenAI-compatible client; default provider DeepSeek (platform.deepseek.com), default
model deepseek-flash (ADR-0008). Both are env-configurable (OPENAI_BASE_URL + AI_DEFAULT_MODEL), so
any OpenAI-compatible provider/model works. (The OPENAI_* env names are just the compatibility
shim — this is not OpenAI/GPT-4.)
Platform — Stripe (payments) · Resend (email) · PostHog (analytics) · Sentry (errors).
A deeper, human-facing architecture overview is in docs/ARCHITECTURE.md.
The exhaustive engineering reference is in CLAUDE.md.
How a summary is built
Frontend ─▶ FastAPI ─▶ SEC EDGAR (filing text + XBRL) ─▶ section extraction
─▶ AI summarization (strict JSON contract, fallback on failure)
─▶ quality assessment ─▶ persist ─▶ streamed back over SSE
Resilience is built in: a circuit breaker and token-bucket rate limiter in front of SEC EDGAR,
JSON-contract repair for model output, and a deterministic fallback summary when the model fails or
times out.
Quick start
# Backend
cd backend && python3 -m venv venv && source venv/bin/activate
pip install -r requirements.txt
cp .env.example .env # set SECRET_KEY and OPENAI_API_KEY
uvicorn main:app --reload --port 8000 # http://localhost:8000/docs
# Frontend (new terminal)
cd frontend && npm install
cp .env.local.example .env.local
npm run dev # http://localhost:3000
Full setup, tests, and workflow: CONTRIBUTING.md.
Project layout
backend/ FastAPI app (routers, services, models, schemas), prompts, evals, tests
frontend/ Next.js app (app/, components/, features/, lib/)
docs/ Architecture, deployment, compliance, troubleshooting
Key API surface (full, always-current reference at /docs):
| Area | Endpoint |
|---|
| Auth | POST /api/auth/register, POST /api/auth/login, GET /api/auth/me |
| Companies & filings | GET /api/companies/search, GET /api/filings/company/{ticker} |
| Summaries | POST /api/summaries/filing/{id}/generate-stream, GET /api/summaries/filing/{id} |
| Multi-Period Analysis (Pro) | POST /api/analysis/{ticker}/stream |
| Subscriptions | POST /api/subscriptions/create-checkout-session |
Deployment
Backend deploys to Google Cloud Run automatically via GitHub Actions on push to main (gated on
tests); the frontend deploys to Vercel via its GitHub integration. See
docs/DEPLOYMENT.md.
Privacy
The maintained privacy notice explains data handling and
available choices. Source review notes are in docs/legal/PRIVACY_POLICY.md.
Contact
Questions, help and feedback: [email protected].
Subscription questions: [email protected].
Privacy requests: [email protected].
Vulnerability reports: [email protected].
See the email setup guide for routing and operations.
License
Proprietary — all rights reserved.
This software is not open source. See LICENSE. No use, copying, modification, or
distribution without prior written permission.
End users of the app/subscriptions are governed by the End-User License Agreement.
