Live & Running Binance Data Feed SaaS · Recurring Revenue India → Global Public Track Record
63Tokens Monitored
58Tradeable Models
55+Min Quality Gate Score
45–75Features Per Model
1hCandle Timeframe
~$30Monthly Infra Cost
Railway Cloud Firebase Firestore Razorpay Payments Binance WebSocket Firebase Auth + JWT XGBoost + Meta-Labeling
The Team

Who Built This

Two engineers. No external funding. Built in full production with live users, real payment processing, and a public track record. No mockups. No waitlists.

AK
Animesh Kukreti
Co-Founder · Lead Engineer

Designed and built the entire AEGIS signal engine: XGBoost training pipeline, meta-labeling architecture, 8-layer signal filter, edge-score quality gate, regime-adaptive thresholds, beginner onboarding layer, FastAPI backend, Firestore integration, and WebSocket real-time delivery. Full-stack from ML to frontend.

animeshkukreti60@gmail.com

AJ
Aman Juiyal
Co-Founder · Product & Growth

Product direction, market positioning, user acquisition strategy, and go-to-market execution. Responsible for the subscription model design, pricing strategy, community building, and India-first user growth. Manages the business side so the engine can focus on signal quality.

Based in Dehradun, Uttarakhand, India  ·  Operated under AEGIS v1.0

The Problem

The Crypto Signal Market Is Broken

Retail crypto signal tools are dominated by two dysfunctional models: Telegram channels where humans manually post calls with 30–90 second delays and zero risk management, and simple indicator bots that fire on a single RSI or MACD crossing with no market context. Both are optimized for engagement, not for your money.

Telegram Channels

  • Manual entry — stale price by the time you act
  • No backtesting, no quantified edge
  • Survivorship bias — losses quietly disappear
  • 20–50 signals/day optimized for “activity”

Simple Indicator Bots

  • One model for all coins — BTC and micro-caps treated identically
  • No S/R context — fires mid-range into noise
  • No expected-value check — negative-EV signals reach users
  • No explainability — users don’t know why a signal fired

The Real Cost

  • Retail traders lose an average of 80% within 12 months
  • Signal quality is never independently audited
  • No education — users don’t improve, they just resubscribe
  • No risk management layer — no SL/TP discipline
Live Proof

The Platform Is Real — Not a Demo

Every metric below is live and verifiable. The track record is public, unedited, and contains both wins and losses. No cherry-picking. No simulated results.

63Tokens scanning live right now
24/7Engine runtime, Railway cloud
~1sLive price via Binance WebSocket
PublicTrack record, every trade shown
3-dayFree trial, no card required
$0External funding raised
View Public Track Record Try the Live Dashboard
Market Timing

Why This Moment Is the Window

Three structural forces are converging in India that didn’t exist 3 years ago and won’t stay open for long.

India’s Retail Crypto Surge

  • 20M+ active traders in India, fastest growth in Asia-Pacific
  • UPI + Razorpay made micro-SaaS subscriptions frictionless
  • Indian retail is under-served by global signal tools (USD pricing, no UPI)
  • Regulatory clarity improving — 30% tax acknowledged = legitimized activity

Global Retail Crypto at Scale

  • 420M+ crypto users globally (Crypto.com 2024)
  • Crypto signal tools market: ~$200M ARR and growing
  • $99/mo incumbents (3Commas, Pionex Pro) vulnerable to price disruption
  • English-speaking Southeast Asia and MENA: next addressable wave

AI Credibility Moment

  • Retail users now trust AI-driven tools in a way they didn’t 2 years ago
  • XGBoost + meta-labeling is institutional methodology now accessible to retail
  • First-mover advantage in quantitative crypto signals at $14/mo price point
  • 63-token fleet with per-token models, beginner onboarding layer, and public track record already live
Technical Architecture

How AEGIS Works — Under the Hood

AEGIS is a quantitative signal engine built on XGBoost with meta-labeling — a technique used by institutional quant funds to filter raw directional predictions into high-precision, tradeable signals.

01

Per-Token XGBoost Primary Models ML Core

Each of 63 monitored tokens has its own XGBoost classifier trained on ~7,000 hours of OHLCV data + 100+ engineered features: ATR, RSI, MACD, Supertrend, Fibonacci, support/resistance, funding rate, open interest, Fear & Greed index, BTC correlation, candlestick patterns, and more. Training uses purged time-series cross-validation — no lookahead leakage, strict temporal ordering with embargo windows.

02

SHAP Feature Pruning Feature Selection

After initial training, SHAP values rank features by cumulative importance. Features accounting for less than 10% of total importance are dropped, reducing from 100+ to 45–75 per model. Each token retains a different feature subset — EGLD needs different signals than BNB.

03

Optuna Hyperparameter Optimization 60 Trials Per Token

60 Bayesian optimization trials per token (max_depth, learning_rate, subsample, gamma, regularization). Objective function is out-of-fold logloss on the purged validation set — optimized for confidence calibration, not accuracy inflation.

04

Meta-Labeling Layer Institutional Technique

A second XGBoost model (the “meta model”) is trained on out-of-fold predictions of the primary model. Its job: given the primary model says BUY — is this a good BUY? Output: a meta-confidence score (0–1). A signal only reaches the user if meta-confidence exceeds a per-token, per-side threshold tuned to achieve ≥62% holdout precision. Below threshold → HOLD, regardless of primary model strength.

05

27-Regime Adaptive Thresholds Regime Gate

A secondary optimizer finds per-regime thresholds on a rolling 600-bar out-of-sample window. Regimes: 3 volume tiers × 3 volatility tiers × 3 trend tiers = 27 buckets. The confidence bar for a BUY in “high volatility, downtrend” differs from “low volatility, uptrend” — automatic adaptation, no manual tuning.

06

8-Layer Signal Filter Signal Gate

Even after meta gate clearance, 8 structural filters run: (1) AI probability & class gap, (2) S/R proximity gate, (3) Reversal pattern scoring, (4) 3-candle confirmation, (5) Confluence vote, (6) Positive expected value required, (7) Macro regime filter, (8) BTC correlation brake.

07

Edge-Score Quality Gate & Beginner Layer Quality + UX

A dual-layer user-protection system added above the model gates. First, an edge-score quality floor (0–100 scale, minimum 55) blocks low-confidence signals that the predictor threshold alone lets through — catching scale mismatches and low-conviction misfires before they reach the virtual wallet or user. Second, a beginner onboarding layer: pre-trade checklist popup on every session open, Guardian playbook with paper-trading-first philosophy, 1–2% risk rule, SL discipline, and a full in-dashboard Risk & Capital Management guide with capital tiers, position sizing formula, and R:R table.

08

Real-Time Delivery Infrastructure

Engine scans all 63 tokens on 1h candles, continuously, on Railway cloud. Live prices stream via Binance’s WebSocket mini-ticker (~1s updates), pushed to dashboard via FastAPI WebSocket. Signals written to Firestore for persistent storage and real-time sync. Frontend receives ticker updates every 100ms, full signal payloads every 500ms. Virtual paper wallet tracks open positions with TP/SL exit logic and syncs to the public track record on every close. Stack: Python 3.12, FastAPI, XGBoost, ccxt, Firebase, Railway.

Business Model

SaaS Subscription — Tiered Access

AEGIS operates on a recurring subscription model with a 3-day free trial (no credit card) as the primary acquisition funnel. Revenue scales directly with subscriber count — near-zero marginal cost per additional user.

Feature Trial (3 days) Basic · $3.60/mo Intermediate · $7.20/mo Pro · $14/mo
Tokens monitored63 (monitor)42 live58 live63 (all)
S/R levels & price targets
Plain-English analysis
AI conviction labelHIGH / MED / LOW
Raw AI confidence %
Fire signal + direction
Guardian + Risk Guide
Multi-timeframe access1h only15m, 30m, 1hAll TFsAll TFs

Unit Economics

  • Basic ARPU: $3.60/mo → $43 LTV/year
  • Pro ARPU: $14/mo → $168 LTV/year
  • Infrastructure: ~$30/mo fixed cost at current scale
  • Break-even: ~3 Pro subscribers
  • Zero marginal cost per additional user

Acquisition Funnel

  • 3-day free trial, no credit card
  • Full platform access during trial — hook on value
  • Razorpay: UPI, cards, netbanking (India-first)
  • Stripe planned for USD/EUR/GBP global billing
  • Referral / affiliate program in roadmap

Target Users

  • Retail crypto traders (India: 20M+ active)
  • Swing & position traders (1h–1w horizon)
  • Beginners wanting guided context, not raw charts
  • Intermediate traders wanting a second opinion
  • Global English-speaking retail market (Phase 2)
Competitive Position

How AEGIS Is Different

Typical Competitors

  • One universal model for all tokens
  • Simple RSI/MACD crossover rules — no ML
  • No backtesting, no holdout quality gate
  • Signals with no explanation — you don’t know why
  • Pricing: $30–$99/month for the same quality
  • No expected value calculation — negative-EV signals sent
  • Static thresholds — no regime adaptation
  • Track record cherry-picked or absent

AEGIS

  • 63 individual XGBoost models — each token trained separately, 58 live
  • Meta-labeling: institutional technique for precision over recall
  • 62%+ holdout precision required before any model goes live
  • Plain-English “Why” analysis on every token, every scan
  • Pricing: $3.60–$14/month — 3–7× cheaper than incumbents
  • Positive EV required — negative-EV signals are killed
  • 27-regime adaptive thresholds per token
  • All closed trades (wins + losses) publicly visible
Honest Assessment

Where We Fall Short — and Our Fixes

We believe the most trustworthy presentation is explicit about current limitations. Each weakness has a concrete mitigation already in development or shipped.

✓ Coverage & Throughput — Fixed

Early versions capped at 6 concurrent virtual positions with a 48-hour hold, limiting closed trade throughput to ~3 per day.

Done: Portfolio guard expanded to 12 concurrent positions (3 per correlation cluster), max hold reduced to 24h, and per-trade position size reduced to 7% base (from 10%) so more positions fit within the 60% capital cap. Expected throughput: 8–12 closed trades per day.

✓ Signal Quality Gate — Fixed

Meta-threshold values in meta files (0.6) were calibrated on the old 0–1 confidence scale, but edge scores are now computed on a 0–100 scale — meaning the predictor's fire gate was effectively never blocking anything.

Done: A hard edge-score quality gate (minimum 55/100) is now enforced at the live engine level before any virtual position opens. Low-confidence misfires (e.g. edge=3.0) are blocked regardless of the predictor's fire=True output.

⚠ Small Live Track Record

The public live track record is growing but not yet statistically significant. Backtesting is rigorous; live performance takes time to compound at scale.

Fix: Transparent, unedited public track record accumulates automatically at 8–12 closed trades per day. All trades (wins and losses) are shown. Statistical confidence will compound over the coming months.

⚠ Signals Only — No Auto-Execution

AEGIS identifies high-probability setups. It does not place orders. Execution, position sizing, and stop discipline are the user’s responsibility — introducing delay and emotional interference. New users are guided to practice on TradingView first.

Fix (Q4 2026): Exchange API integration for optional one-click execution via Binance and Bybit. Users retain full control; auto-execution is opt-in.

⚠ Model Drift Risk

XGBoost models are trained on historical data. In genuinely unprecedented conditions (rapid regulatory change, black swan events), a model’s patterns may degrade before the next scheduled retrain.

Fix (Q3 2026): Automated drift detection. If live precision diverges from backtested expectation, a retrain triggers automatically within 24 hours.

⚠ India-First Payment Infrastructure

Current payment processing (Razorpay) is optimized for Indian users. International users face friction: no native USD/EUR billing, limited card acceptance outside India.

Fix (Q4 2026): Stripe integration for USD/EUR/GBP billing. Global expansion begins with English-speaking crypto markets: UK, US, Southeast Asia, MENA.

Product Roadmap

Where AEGIS Is Going

Live Now

Core Signal Engine — 60 Tokens, 24/7

All 60 tokens with live prices, RSI, trend regime, funding rate, plain-English analysis. 24 tradeable models. Real-time WebSocket. Public track record.

Live Now

AEGIS Trader Bot — Universal AI (Scalping / Intraday / Swing)

Universal XGBoost trained on 10 tokens, deployed on all 60. 25 strategies + 20 feature-engine indicators. 3 modes × 3 risk profiles. Virtual paper wallet with real trade history. Separate track record.

Q3 2026

Automated Model Drift Detection + Retrain

Monitors live precision vs backtested expectation. Triggers automatic retrain when drift exceeds threshold. Zero manual intervention required.

Q4 2026

Broker Integration — Optional Auto-Execution

API connection to Binance and Bybit (opt-in). AEGIS places orders automatically when signals fire. Removes execution delay entirely.

Q4 2026

Stripe Global Billing + International Expansion

USD/EUR/GBP billing via Stripe. English-speaking markets: UK, US, Southeast Asia, MENA. Affiliate / referral revenue share program.

2027

On-Chain Sentiment Layer + Portfolio Risk View

Wallet flow data, exchange inflows/outflows, social sentiment as live model features. Portfolio risk dashboard: total exposure, cross-token correlation, combined max drawdown.

Market Opportunity

  • 20M+ active crypto traders in India alone
  • 420M+ crypto users globally (Crypto.com, 2024)
  • Crypto signal tools market: ~$200M ARR, growing
  • $99/mo incumbents leave room for disruption at $14/mo

Technology Stack

  • Python 3.12 — signal engine, ML pipeline
  • XGBoost — primary + meta + trader models
  • FastAPI + WebSocket — real-time API
  • Firebase Firestore — storage, auth
  • Railway — 24/7 cloud deployment
  • Binance WebSocket — live price feed
  • Razorpay → Stripe — payment processing

Current Scale

  • 60 tokens monitored simultaneously
  • 100+ features per token per bar
  • ~7,000 hours training data per model
  • 5-minute scan cycle, ~1s live prices
  • Infrastructure cost: ~$30/mo at current scale
Get In Touch

The Thesis

The retail crypto signal market is large, underserved, and dominated by low-quality products. AEGIS applies institutional quantitative methods — meta-labeling, per-asset models, regime-adaptive thresholds, rigorous backtesting — at a price point any retail trader can afford. The business is capital-light, recurring-revenue, and scales at near-zero marginal cost. The product is live, real, and the track record is public. Every trade. Every loss. No filters.

⚡ Start 3-Day Trial View Live Track Record →
Investor & Partnership Inquiries
Animesh Kukreti
Co-Founder & Lead Engineer
animeshkukreti60@gmail.com
Business email
AEGIS v1.0
animeshkukreti@aegisignal.pro
Dehradun, Uttarakhand, India  ·  aegisignal.pro