Project Case Study
Quantiveo
LiveCrypto Data Intelligence & Decision-Assistant Platform
Quantiveo is a platform for collecting, normalizing, and analyzing market, on-chain, DeFi, sentiment, and whale movement data. It uses a deterministic signal engine and AI narration layer to turn fragmented crypto data into understandable insights for holders, swing traders, and daily traders.
Technical Snapshot
Project Type
Crypto data, analytics, and decision-assistant platform
Anarchain Role
Data architecture, signal engine, Go backend, dashboard, API, backtesting, AI narration
Status
🚀 Live — quantiveo.ai
Target Users
Holders, swing traders, daily traders, analysts, data teams
Architecture
Data Pipeline + Signal Engine + AI Narration
Core Value
Turning fragmented crypto market data into usable intelligence
The Problem
The Problem This Project Solves
Crypto market data is scattered across multiple sources: price and volume in one service, on-chain data in another, TVL and DeFi data elsewhere, social sentiment in another source, and whale transactions in a separate system. Traders and analysts often need to check several dashboards manually, interpret signals themselves, and still end up with an incomplete view.
Data scattered across multiple independent sources
No unified scoring model across metrics
Difficulty comparing assets across different criteria
High noise in short-term data streams
Holders, swing traders, and daily traders need different analysis types
Cost and complexity of subscribing to multiple analytics services
No simple, understandable narrative for non-professional users
No structured dataset for backtesting and future modelling
Data Flow Architecture
Data Flow Topology
In this system, users do not directly read from external services. A dedicated ingestion and scheduling layer collects data from multiple sources, then normalizes, stores, processes, scores, and finally delivers it as understandable analysis to the user.
Technical Approach
Anarchain's Solution
The solution is a multi-layered platform that collects crypto data from multiple sources, converts it to standardized metrics, stores it in appropriate databases, analyzes it with a deterministic signal engine, and finally uses AI to make the output understandable for each user type.
- Collect data from multiple authoritative sources
- Convert data to standardized canonical metrics
- Time-series storage for analysis and backtesting
- Deterministic scoring engine — not random AI output
- AI narration to simplify analysis
- Different output per user type: holder, swing, daily trader
Data Sources
Data Sources
Each data source plays a different role in the analysis. Some sources provide price and volume data, others on-chain data, DeFi metrics, sentiment, or large transaction tracking.
CoinGecko
Market DataPrice, market cap, volume, OHLC, supply data
DeFiLlama
DeFiTVL, DEX volume, fees, yields, stablecoin data
Santiment
SentimentSocial sentiment, developer activity, active addresses, MVRV, NVT
CryptoQuant
On-chainExchange netflow, reserves, whale ratio, funding, SOPR/NUPL
Whale Alert
Whale TrackingLarge transactions, mint/burn, whale movements
Fear & Greed
Context IndexOverall market fear and greed context index
CEX/DEX APIs
Market APIsReal-time price, liquidity, spread, order and market context
User Personas
Persona-Based Intelligence
Not all users need the same type of analysis. Holders care about long-term trends and network fundamentals, swing traders need momentum and market flow data, and daily traders depend more on fast data, volume, price, and whale movements.
Holder
Key Metrics
Output
Long-term analysis, overall risk, accumulation or distribution status
Swing Trader
Key Metrics
Output
Multi-day to multi-week analysis, probable setups, trend reversal signals
Daily Trader
Key Metrics
Output
Alerts, real-time status, rapid changes, short-term risk assessment
System Brain
Signal Engine — The Deterministic Brain
In this project, AI is not responsible for calculating signals. The core calculation happens in a deterministic engine to ensure repeatable, backtestable, and trustworthy output. This engine receives normalized data, applies metric weights based on user type and asset, calculates data freshness, builds confidence, and converts output into understandable signal bands.
- 1Receive normalized metrics
- 2Calculate z-score / relative changes
- 3Apply weights based on user profile and asset type
- 4Calculate data freshness
- 5Calculate confidence score
- 6Determine signal band
- 7Extract primary signal drivers
- 8Store snapshot for backtesting
AI Layer
AI Narration Layer
In this architecture, AI does not replace the analysis engine — it narrates the deterministic engine's output for the user. Numbers, scores, and signals are calculated by the engine, and AI simply explains them in plain language tailored to the user type.
// Narration Output Examples
Overall network conditions are in a positive range, but sentiment data does not yet provide full confirmation.
Short-term volume and large transactions have increased, but confidence is moderate — price confirmation is still needed.
// AI Narration Cycle
System Architecture
Technical Architecture
This platform is designed with a multi-layer architecture to control API costs, keep data backtestable, maintain repeatable calculations, and deliver fast, understandable outputs for different user types.
Backtesting & Data
Backtesting & Future Dataset
One of the core goals of this system is building a structured dataset of market data and signal engine outputs. By storing snapshots, metric points, and narrations, the platform enables backtesting of decisions, weight optimization, and eventually training a custom model.
- Store raw data for replay in S3/MinIO
- Store metric points in TimescaleDB
- Store signal snapshots with timestamps
- Compare signals against subsequent price behavior
- Optimize weights using historical data
- Prepare dataset for a future custom model
// Backtesting Goals
Development Plan
Product Roadmap
MVP Data Core
Connect data sources, store metrics, design data schema
Signal Engine
Weighting, scoring, freshness, confidence, signal bands
Dashboard
Asset views, market status, primary drivers, per-persona output
Backtesting
Store snapshots, replay, compare signals to future price data
AI Narration
Persona-specific narration for holder, swing trader, daily trader
Alerts & Subscription
Alerts, watchlists, periodic reports, subscription plans
Mobile App / API
Mobile app and API for professional users or data teams
Strategic Importance
Why This Project Matters for Anarchain
This project demonstrates that Anarchain doesn't just build websites or simple dashboards — it can design and execute multi-source, backtestable, AI-ready data systems. Crypto Data & Decision Assistant can become a standalone product, educational tool, content source for Coinbazan, and a powerful proof point for attracting data and AI projects.
// Key Points
- Proof of Anarchain's capability in AI/Data Engineering
- Direct connection to the crypto and Web3 market
- Content production foundation for Coinbazan
- Technical proof for attracting data and dashboard projects
- Potential product for a future subscription model
- Educational tool for Anarchain Academy
Work with Anarchain
Need a Data Platform or Decision Assistant System?
If you need a data dashboard, market analytics system, signal engine, AI platform, or decision-assistant infrastructure, Anarchain can support you from architecture and development to backtesting and deployment.