Beomgyu Joeng [Jaden]

BEOM-GYU JOENG [JADEN]

AI‑Native Quant Systems Architect & Agentic‑Automation Trading Engineer

AI‑driven Quant Engineer building institutional‑grade crypto trading systems with +44% verified performance, millisecond ML inference, and agentic automation.

Scroll

My Expertise

Maximize Creativity and Intellectual Potential by Leveraging AI/Agents to Seek Alpha Through Revealing Patterns in Quantitative Trading.

Trading System
Event Driven Architecture
Signal ML/RL Real-time Streaming
HFT (Feasible)
Nautilus Trader (Rust/Python)
ByBit(+), Order Execution
TP/SL (Take Profit/Stop Loss)
Circuit Breaker
Backtesting System
WFV (Walk Forward Validation)
Resilient Model Loading System
Feature Pipeline & Signal Generation
4-tier position sizing Cascade
Regime Detection (HMM)
Infrastructure & DevOps
Kubernetes, Docker & Docker Compose
GHCR Container Registry
PostgreSQL, Timescale DB
Redis (Pub/Sub & Caching)
Azure Blob Storage
Prometheus, Grafana, Loki
Github Actions CI/CD
HelmCharts, Terraform, Ansible
ML Ops (MLflow, Minio)
Azure VMs, VMSS, Linux OS
Quantitative Finance
Option Pricing (Black-Sholes, Merton, Heston, Bates)
Evaluation Metrics (IC, Hit Rate, P-value)
Stochastic Volatility Models
Market Microstructure
Kelly Criterion & Position Sizing
Portfolio Optimization
Risk Management (VaR, CVaR)
Derivatives Pricing
MS-GARCH Regime Models
AI/ML Engineering
Feature Engineering (Stationarity), Boruta
ML Trading (RF, XGBoost, Ensemble)
Scikit Learn, Torch (CPU/GPU)
Transfer Learning
Reinforcement Learning (PPO/SAC)
Deep Learning (LSTM, RNN, Transformer)
Jupyter Notebook, MLflow
Agentic Workflows (Language Model, Multi-Agent)
Agent Based Workflow (Claude Code, Antigravity)
Business Intelligence
Performance Metrics (Sharpe, Sortino, K-Ratio)
Custom Metrics (JADE)
Strategic Planning
Project Management
Cost Optimization
Start-up Ventures
Workflow Automation
System Optimization
Research & Documentation
Vim, LaTeX, Markdown
DDD (Document-Driven Development)
Crisp-DM
Knowledge Graph (Mermaid, LMWPF)
Literature Review & Synthesis
Deep Research based Multi Agents Orchestration
Stochastic Calculus (Ito, HJB)

Projects

A selection of my work, from full-stack trading systems to research prototypes.

Trade-Matrix: Institutional-Level ML/RL Based Automated Trading System In Crypto
Production-deployed crypto trading system with millisecond ML inference, RL position sizing, and 4-tier fallback cascade. Kubernetes deployment with 100+ monitoring metrics on Azure VM and $0/month automation via GitHub Actions.
Azure VM + VMSS
Rust + Python
Redis + PostgreSQL + TimescaleDB
MLflow + MinIO
Kubernetes + Docker
Prometheus + Grafana
View Details
Trade-Matrix: Institutional-Level ML/RL Based Automated Trading System In Crypto
LMWPF: Language Model Work Process Framework
LMWPF enhances Agentic AI for Workflow Automation which can automate: quant research, backtesting, documentation(300+), model evaluation, model deployment etc. It efficiently reduces AI hallucination, context loss, token usage in production by leveraging CLI tools like Claude Code.
Session Continuity Protocol
Session Knowledge Graphs
Multi-Agent Orchestration by DAAO
Workflow Management
Document-Driven Development
View Details
LMWPF: Language Model Work Process Framework
Quantitative Finance Deep Research
1000+ pages of deep research across 6 domains with 150+ formal equations and 200+ citations. Production-validated implementations in market making, options pricing, ML algorithms, Regime detection and RL position sizing.
Machine Learning (RF, XGBoost, Enseble)
Deep Learning (Transformers, LSTM)
Reinforcement Learning (PPO, SAC)
Options Pricing Models
Market Microstructure Theory
View Details
Quantitative Finance Deep Research

Research & Ideas

A collection of my articles, explorations, and ideas on finance and technology.

All
Regime Detection
System Architecture
Risk Management
Quantitative Finance
Deep Learning
Options
Market Making & DeFi
Cryptocurrency
Arbitrage
Transfer Learning
Machine Learning
Walk-Forward Validation
Ensemble Methods
Production ML
Algorithmic Trading
Backtesting
Volatility Modeling
Data Science
Transformers
Feature Engineering
Bayesian ML
Reinforcement Learning
LLM Architecture
Multi-Agent Systems
AI Agents
High Performance Computing
MS-GARCH Enhancement Report: From Research to Production
Regime Detection
System Architecture
Risk Management
MS-GARCH Enhancement Report: From Research to Production

January 24, 2026

Updated: Jan 26, 2026

Comprehensive synthesis of MS-GARCH research (Oct 2025 - Jan 2026) with 6 key lessons learned and implementation roadmap for Trade-Matrix regime detection improvements. Includes code examples and quantitative validation.
Read More
Option Pricing Models for Cryptocurrency Derivatives
Quantitative Finance
Deep Learning
Options
Option Pricing Models for Cryptocurrency Derivatives

December 10, 2025

Updated: Jan 8, 2026

From Black-Scholes to Bates: stochastic volatility models calibrated for crypto markets, with neural network acceleration achieving 20-60x speedup and 1.8% RMSE accuracy.
Read More
Market Making for Algorithmic Trading
Unrevealed
Market Making & DeFi
Cryptocurrency
Arbitrage
Market Making for Algorithmic Trading

December 5, 2025

Updated: Jan 9, 2026

Comprehensive survey of market making theory from traditional Glosten-Milgrom to DeFi AMMs, covering inventory management, optimal quoting, and cross-exchange arbitrage strategies.
Access Restricted
Transfer Learning for Financial Time Series
Transfer Learning
Machine Learning
Walk-Forward Validation
Ensemble Methods
Production ML
Transfer Learning for Financial Time Series

November 15, 2025

Updated: Jan 25, 2026

A production-validated 3-phase Transfer Learning protocol for cryptocurrency signal prediction, achieving weekly model updates in ~60 minutes (full pipeline, ~20 min training step) while preventing catastrophic forgetting.
Read More
MS-GARCH Weekly Frequency Optimization
Interactive
Regime Detection
Algorithmic Trading
Cryptocurrency
MS-GARCH Weekly Frequency Optimization

October 15, 2025

Updated: Jan 24, 2026

Validating weekly (1W) frequency for MS-GARCH regime detection. Achieves 8.38× longer regime durations (27 days vs 3.26 days daily), 89% fewer transactions (13/year vs 112/year), and 11.8% transaction cost savings.
Explore Notebook
MS-GARCH Backtesting Validation: Walk-Forward Framework
Interactive
Regime Detection
Risk Management
Backtesting
MS-GARCH Backtesting Validation: Walk-Forward Framework

October 12, 2025

Updated: Jan 24, 2026

Economic validation of MS-GARCH regime detection through walk-forward backtesting. Regime-conditional leverage achieves Sharpe 1.69 (Moderate) with systematic validation via Kupiec POF VaR test.
Explore Notebook
MS-GARCH Model Development: 2-Regime GJR-GARCH for Cryptocurrency
Interactive
Regime Detection
Volatility Modeling
Cryptocurrency
MS-GARCH Model Development: 2-Regime GJR-GARCH for Cryptocurrency

October 11, 2025

Updated: Jan 24, 2026

Implementing Markov-Switching GARCH models for volatility regime detection in cryptocurrency markets. BIC-optimal 2-regime model achieves 74% Low-Vol / 26% High-Vol classification with 5.1-week average persistence.
Explore Notebook
MS-GARCH Data Exploration: CRISP-DM Data Understanding
Interactive
Regime Detection
Data Science
Cryptocurrency
MS-GARCH Data Exploration: CRISP-DM Data Understanding

October 10, 2025

Updated: Jan 24, 2026

Implementing the Data Understanding phase of CRISP-DM methodology for MS-GARCH regime detection. Analysis of cryptocurrency volatility patterns, return distributions, and stylized facts for BTC, ETH, and SOL.
Explore Notebook
Advanced Machine Learning Algorithms for Trading
Machine Learning
Deep Learning
Transformers
Ensemble Methods
Feature Engineering
Bayesian ML
Advanced Machine Learning Algorithms for Trading

October 1, 2025

Updated: Jan 8, 2026

Comprehensive survey of ML architectures for financial time series, from gradient boosting ensembles to Temporal Fusion Transformers, with benchmark comparisons and implementation roadmap.
Read More
Reinforcement Learning for Optimal Position Sizing
Reinforcement Learning
Risk Management
Algorithmic Trading
Reinforcement Learning for Optimal Position Sizing

September 20, 2025

Updated: Jan 25, 2026

A Kelly-convergent Soft Actor-Critic framework with 4-tier fallback cascade for robust position sizing in weak signal regimes, achieving 45-minute training through curriculum learning.
Read More
Hidden Markov Models for Market Regime Detection
Regime Detection
Risk Management
Quantitative Finance
Hidden Markov Models for Market Regime Detection

September 1, 2025

Updated: Jan 26, 2026

MS-GARCH implementation for cryptocurrency regime detection using 4-state Hidden Markov Models, enabling regime-adaptive position sizing from 25% (Bear) to 67% (Bull).
Read More
Session Continuity in Large Language Model Systems
Unrevealed
LLM Architecture
System Architecture
Session Continuity in Large Language Model Systems

August 31, 2025

Addressing the fundamental challenge of knowledge persistence across context-limited LLM interactions through the Session Continuity Protocol (SCP).
Access Restricted
Multi-Agent Orchestration for Complex Knowledge Work
Unrevealed
Multi-Agent Systems
System Architecture
AI Agents
Multi-Agent Orchestration for Complex Knowledge Work

August 31, 2025

Designing intelligent agent systems with specialized roles, difficulty-aware task routing, and PM-style supervision for reliable autonomous operation.
Access Restricted
High-Performance Technical Analysis with TA-Numba
High Performance Computing
Algorithmic Trading
High-Performance Technical Analysis with TA-Numba

July 6, 2025

TA-Numba is a Python library for financial technical analysis that provides dependency-free installation and high-performance computation through Numba JIT compilation. It offers both bulk processing for historical analysis and real-time streaming for live trading applications.
Read More
GitHub

Thoughts

Reflections on technology, finance, and the evolving landscape of algorithmic trading.

2026-01-18

The Multiplicative Era: How Agentic AI Will Reshape Knowledge Work in 2026

Timeline

45 activities in 2026

JanFebMarAprMayJunJulAugSepOctNovDec
W5W10W15W20W25W30W35W40W45W50
Less
More

Activity

trade-matrixDoneJan 27
K3S Deployment: VVIX T-1 Fix

Deployed external data alignment fix to production, achieved exact signal parity between Docker and K3S

trade-matrixDoneJan 27
External Data Integration Guidelines

New comprehensive guide (2,219 lines) for external data integration based on VVIX lessons learned

trade-matrixDoneJan 27
LMWPF User Manual v1.9.6

Updated LaTeX manual with T-1 Temporal Alignment pattern, code examples, and workflow diagrams

Knowledge Map

An interactive visualization of my skills, domains of expertise, and the tools I use. Hover over a node to see how my knowledge connects.

Work Experience

My professional journey in quantitative finance and technology.

Systematic Trader

Top 2%

Nov 2025 (3 weeks)

TradingView Trading Competition

  • Achieved top 2% ranking out of 82,000 competitors in TradingView "The Leap" competition
  • Generated 25.55% return ($100K to $125.5K) using systematic futures trading strategies
  • Demonstrated exceptional risk management with minimal drawdown during volatile market conditions
  • Traded CME futures (E-mini Nasdaq, Bitcoin, Ethereum) with high Sharpe ratio
Systematic Trading
CME Futures
Risk Management
Technical Analysis
View Full Competition Details

Founder & Lead Quant Systems Architect

Jan 2025 - Present

Trade Matrix Labs

  • Founded algorithmic trading startup developing institutional-grade cryptocurrency trading systems
  • Designed and deployed a fully automated trading system as a solo engineer, including ML signal engine, RL position sizing, execution layer, and monitoring stack
  • Built comprehensive research portfolio spanning market making, options pricing, and quantitative strategies
  • Developed LMWPF framework for AI-assisted development with 42% efficiency improvement
Start-up Ventures
Automated Trading
Python
Machine Learning
Reinforcement Learning
NautilusTrader
Kubernetes
MLflow
PostgreSQL
Redis

RLHF Expert

Sep 2024 - Jan 2025 (5 months)

Outlier

  • Contributed to Reinforcement Learning from Human Feedback (RLHF) for AI model training
  • Specialized in mathematical reasoning and coding evaluation (Python, Java)
  • Provided expert feedback to improve AI model accuracy in quantitative domains
  • Applied rigorous mathematical analysis to validate model outputs
Applied Mathematics
Python
Java
RLHF
AI Training

Co-Founder / Tech Lead

Aug 2018 - Sep 2023 (5 yrs 2 mos)

ZAMMA GmbH / Jin Entertainment

  • Co-founded and led technology development for entertainment and events company
  • Built and managed technical infrastructure for large-scale cultural events
  • Applied business analytics to optimize operations and growth strategies
  • Led KOREAN NIGHT | JIN ENTERTAINMENT event series production
Start-up Ventures
Business Analytics
Event Technology
Project Management

Education

Academic foundation in finance, technology, and quantitative methods.

WorldQuant University

Jan 2025 - May 2026 (Current)

Master of Science - MS, Financial Engineering

Grade: 4.0/4.0

Financial Analysis
Quantitative Analytics
Derivatives Pricing
Portfolio Optimization
Risk Management

Technical University of Munich

Oct 2017 - Mar 2024

Bachelor's degree, Management and Technology

Specialized in Computer Engineering

FastAPI
PostgreSQL
Python
Machine Learning
System Design
Algorithms
Data Structures
Web Development

Volunteering & Training Programs

Professional development through industry-leading training programs and initiatives.

Google Cloud Skills Boost

Trainee at Google

Mar - Sep 2024 (7 months)

Advanced training in Generative AI and Google Cloud Platform with focus on LLM operations and production deployment

Education
Generative AI
Vertex AI
Gemini
PaLM API
Google Cloud Platform
LLMOps

Google Machine Learning Bootcamp 2023

Trainee at Google Developers

Aug - Dec 2023 (5 months)

Comprehensive bootcamp covering machine learning and deep learning fundamentals with hands-on projects

Science and Technology
Machine Learning
Deep Learning
TensorFlow
Neural Networks
Computer Vision
NLP

Get In Touch

Have a question or want to work together? Drop me a line.

Contact Information

Open to relocation to Dubai/Singapore for Quant/AI-Native System Architect/Crypto Trading Engineer roles.

jaden.b.jeong@gmail.comGitHubLinkedInX.com