Quantitative Researcher | High-Frequency Trading Enthusiast | Algorithm Developer | Agent Developer
I am passionate about leveraging statistical modeling, machine learning, and low-latency system design to solve complex financial problems. My focus lies at the intersection of mathematics, computer science, and alpha generation.
yang13515360252@163.com
- Languages:
C++(Low-latency/Modern C++),Python(Data Science),SQL(ClickHouse) - Libraries/Frameworks:
Eigen,OpenMP,STL,NumPy,Pandas,light-gbm,PyTorch,langchain,langgraph - Infrastructure & Tools:
Linux,Git,Docker - Math & Finance: Stochastic Analysis, Time Series Analysis, Convex Optimization, Market Microstructure
[Beijing Key Laboratory of Financial Artificial Intelligence] | Core Researcher | Apr 2025 – Mar 2026
-
Built multi-agent LLM systems for banking operations, integrating internal data warehouses and external platforms; designed regulatory knowledge graphs, tool APIs, and fine-tuning QA datasets; fixing vulnerabilities in dependencies detected by Fortify scan.
-
Developed CUFEL-A, a multi-agent framework for automated A-share research report generation with a paper-planning–architecture-report structure , web search tools, and prompt optimization.
-
Built CUFEL-Q Arena, a multi-agent quantitative trading platform with multi-frequency financial data pipelines, NLP/TDA text indicators, and high-precision backtesting engines.
-
Developed physics-informed deep learning models for Tokamak nuclear fusion, combining TimeXer encoders, CNN/FNN decoders, and autoregressive forecasting.
[Minghuhui Private Investment Fund] | High-Frequency Quantitative Research Intern | Aug 2024 – Jan 2025
-
Built vectorized high-frequency backtesting engines in Python; accelerated intraday simulation via matrix operations and Dask parallelization.
-
Constructed 100+ LOB and trade-flow factors using Python + ClickHouse; evaluated predictive power via IC/ICIR.
-
Developed DeepLOB and LightGBM trading models for convertible bonds and BTCUSDT T0 strategies.
[Huatai Securities] | Quantitative Research Intern | Aug 2023 – Apr 2024
-
Developed a Python backtesting framework for equity strategies with portfolio construction and IC/RankIC evaluation.
-
Implemented analyst forecast surprise strategies for A-share equities and conducted empirical backtests.
-
Built tools for exchange-traded option strategy analysis, including implied volatility estimation and Greeks evaluation.
📈 1. GPFactor
- Description: Factor Research Series 1: Genetic Programming Factor Mining Algorithm implemented in C++.
- Tech:
C++20,OpenMP,Eigen
📈 2. MAS-FactorMiner
- Description: Factor Research Series 2: LLM-Driven Multi-Agent System for Explainable Alpha Discovery.
- Tech:
langchain,function call
⚡ 3. GeneralBacktest
- Description: A General Backtest Framework for Quantative Research.
- Tech:
python,numpy,pandas
- Description: A ETF Strategy based on PTree data selection for stable performances.
- Tech:
PTree,iTransformer
- Description: This project explores the time-series predictability of the HS300 ETF by leveraging a "zoo" of 105 high-frequency characteristic-sorted portfolios.
- Tech:
jump-diffusion decomposition,high-frequency trading
- 🔬 [QuantMoE Model]: Mixture of Experts DL Models for stocks trading with a specified optimization function.
- 📚 [Market Regimes Detector]: Using Reinforcement Learning and Markov probability transition matrix to predict Market Regimes.
- 🧠 [Hybrid Data Factor Miner]: Using multi-frequency trading data and nlp indicator to discover new factors.
