Contact person: Enterprise Financial Analytics Platform

Phone:Show

Email:Show

Location: Riyadh, United States

Budget: Recommended by industry experts

Time to start: As soon as possible

Project description:
"Financial Intelligence Web App — Full Project Requirements & Deliverables
1) Project Overview
Build a production-grade web platform similar to [login to view URL] or google financial for U.S. public companies and others, featuring:
• Fundamentals dashboards (Balance Sheet, Income Statement, Cash Flow, ratios, Key Metrics, prices).
• A similarity/vector search engine over multi-year financial vectors (incl. DTW and classic metrics, yearly and quarterly).
• Advanced screeners (rule-based + vector-based).
• SEC Filings RAG (Q&A over 10-K/10-Q/20-F/40-F with chunk references).
• Natural-language to Cypher (N2C) for querying Neo4j.
• Admin panel to run/update ETL pipelines from financial data source API and configurable endpoints.
• User accounts, role-based access, and subscription tiers (Free/Pro/Enterprise).
Target user: enterprises /analysts /investors.
Data backbone: Neo4j (20–30 years of history), enriched via financial data source API
________________________________________
2) Goals & Success Criteria
• Accuracy: Numeric parity with financial API for ingested fields; deterministic normalization; clear audit logs.
• Speed: P95 < 1.5s for typical dashboard queries; < 4s for similarity queries (CPU).
• Usability: Cohesive UI, clear data provenance, downloadable tables/charts, and shareable links.
• Explainability: Every similarity result has feature-attribution & time-window disclosure.
• Reliability: Zero-data-loss ETL with resumable jobs and idempotent writes.
• Security: Role-based access; secrets isolated; least-privilege.
• Monetization: Subscriptions integrated with Stripe.
________________________________________
3) Scope (What to Build)
A. Data & ETL Layer
• ETL orchestrator in Python to ingest data from API.
• Handles annual & quarterly data, 25+ years backfill, retries, JSON logs, and CSV audit. (first run)
• Daily update new data if exists.
• Graph schema in Neo4j with nodes for Company, FS, Ratios, Key Metrics, Prices, etc (about 20 API endpoints).
• Vectors per year: ratios, metrics and composites (configurable).
B. Vector & Similarity Analytics
• Multiple similarity methods: cosine, Euclidean, Pearson, Mahalanobis, DTW.
• Supports composite vectors and explainability (feature attribution).
• GDS clustering per year (Leiden, Label Propagation) with undirected edges.
C. SEC Filings RAG (Q&A and pre-defined questions)
• Filings fetcher, parser, and chunker (HTML/txt).
• Supports Q&A with citation references to source filings.
• Handles numeric tables and textual disclosures.
D. Natural-Language to Cypher (N2C)
• Converts user questions to Cypher queries over Neo4j by LLM.
• Safe schema-aware execution with fallback strategy.
• Get the query results explanation by LLM.
E. Web App (Front End)
• Pages:
1. Landing page: summary of stock market
2. Company Overview:
 Tabs:
 Financials (BS/IS/CF with annual/quarterly toggle)
 Ratios & Key Metrics
 Analyst estimates
 Similarity Search
 News
3. Screeners (rule-based + vector-based)
4. SEC RAG Q&A: based latest SEC filing.
5. Admin Panel
6. User Account & Subscription Management
• Features: CSV/Excel export, watchlists, saved screeners, light/dark mode.
F. Admin & Ops
• Job runner for ETL tasks with live logs.
• Config editor for API endpoints params & weights.
• User & subscription management.
________________________________________
4) Database & Users
Database
• Neo4j: main graph DB (dev/staging/prod environments).
• Daily backup/restore procedures.
• Indices & constraints for query performance.
• Redis caching for frequent queries.
Users & Authentication
• Auth system:
o Email/password (JWT-based).
o Google OAuth (Phase 2).
• Roles:
o Admin: ETL jobs, configs, billing.
o Analyst: full access to data & similarity.
o Viewer: dashboards only.
• User profiles: saved searches, watchlists, API quota tracking.
________________________________________
5) Subscriptions & Monetization
• Free Tier:
o Limited company lookups per day.
o Dashboards only.
• Pro Tier:
o Unlimited dashboards.
o Full similarity + screeners.
o Unlimited SEC RAG queries/month.
• Stripe Integration:
o Monthly/annual billing.
o Auto-renewal & cancellation management.
o Admin revenue dashboard.
• Access Control:
o API gateway enforces subscription limits.
o Feature flags toggle advanced features by tier.
________________________________________
6) Non-Functional Requirements
• Stack: Python FastAPI, Neo4j, Redis, React/[login to view URL], Tailwind, Docker.
• Performance: P95 <1.5s dashboards; <4s similarity queries.
• Observability: JSON logs, Prometheus metrics, Grafana dashboards.
• Security: Role-based access, OWASP compliance, secrets in env.
• Compliance: financial API data source ToS & rate limits respected.
________________________________________
7) Deliverables
• Full codebase (ETL, API, analytics, web, infra).
• Neo4j schema & migration scripts.
• Vector builder with similarity methods.
• Web UI with dashboards, similarity, screeners, RAG, subscriptions.
• Admin panel with ETL jobs & user management.
• Stripe billing integration.
• Tests (unit/integration/load).
• Documentation (schema diagrams, runbooks, API docs).
• Deployment scripts (Docker, CI/CD).
________________________________________
8) Milestones
1. M1 – Architecture & Schema
2. M2 – ETL Foundation
3. M3 – Vectors & Similarity
4. M4 – Web App v1 (Dashboards, Screeners)
5. M5 – RAG + N2C
6. M6 – Subscriptions + Admin Panel
7. M7 – Final Testing & Handover
________________________________________
9) Developer Qualifications
• Python (ETL, FastAPI), Neo4j + GDS.
• Time-series similarity (DTW, trajectory analysis).
• React/[login to view URL] with financial dashboards.
• RAG for long-text docs (HTML).
• Payment integration (Stripe).
• Docker + CI/CD + cloud deploy.
________________________________________
10) What You Provide in Proposal
• Architecture plan with diagrams.
• Tech stack confirmation (or suggested changes).
• Examples of similar projects.
• Draft timeline with milestones.
• Risk/assumptions & mitigation plan." (client-provided description)