Home Quantitative Machine Learning Internship at BOC International

Quantitative Machine Learning Internship at BOC International

Quantitative Machine Learning Internship at BOC International

Fintech Department

From 04-2023 to 07-2023

During my time as a Quantitative Machine Learning Intern in the bustling financial hub of Shanghai, China, within a dynamic Fintech Department from April to July 2023, I embarked on a deep-dive into the fusion of advanced machine learning and high-frequency trading (HFT).


  • Design, training, optimization, deployment and trading algorithm design, development and optimization of time series based Transformer, GPT, BERT architecture for high frequency trading models.

  • Analyze large amounts of financial market data, discover and interpret patterns, and work with a team of software engineers to embed the models into trading systems.

Leveraging Time Series and Transformer Models:

My core mission was to design and develop cutting-edge trading algorithms. I focused on the intricacies of time series analysis, employing Transformer architectures—reminiscent of the celebrated GPT and BERT models, renowned for their proficiency in understanding sequences. This endeavor wasn’t just about theory but about creating models robust and responsive enough to handle the frenetic pace of high-frequency trading environments.

From Conceptualization to Trading Floor:

The journey of an algorithm from a mere concept to deployment involved several critical stages:

1.Design and Training:

I conceptualized and trained sophisticated machine learning models, tailoring the architectures specifically for the nuances of financial time series data. This stage was all about capturing the patterns that traditional analysis could miss, learning from the vast troves of past market behavior.


A model is only as good as its performance. Hence, I iteratively refined these models, fine-tuning hyperparameters and structures to squeeze out every ounce of predictive power, all while rigorously ensuring they avoided overfitting to past data.


The transition from a research environment to a live market was where theory met reality. Deploying the optimized models into the trading infrastructure required not just machine learning expertise but also a solid understanding of software engineering practices and collaboration with seasoned software engineers.

4.Trading Algorithm Development and Optimization:

Post-deployment, my focus shifted to the continual improvement of the trading algorithms. This entailed developing new strategies and refining existing ones to adapt to the evolving market conditions.

Data Analysis and Team Collaboration:

A significant part of my role involved analyzing massive datasets of financial market information. By identifying and interpreting complex patterns within the data, I extracted actionable insights. Collaborating closely with a team of skilled software engineers, I worked to integrate the predictive models into the firm’s trading systems seamlessly.

Throughout this experience, my proficiency in machine learning and quantitative analysis was not just applied but tested and honed in the real-world cauldron of financial trading, an arena where milliseconds can mean the difference between profit and loss. This internship was a vivid testament to the transformative potential of AI in the financial sector, particularly in the demanding world of high-frequency trading.

This post is licensed under CC BY 4.0 by the author.

Publish Markdown to Medium in 2 Minutes

Machine Learning Engineer Internship at Shengqu Games

Comments powered by Disqus.