Open to opportunities

Ben Liu

@benliu

Message

Quant developer and data engineer specializing in energy markets, streaming analytics, and quantitative modeling.

United States

Message

What I'm looking for

I seek engineering roles where I can build scalable data and analytics platforms, contribute to quantitative modeling for trading, and work with cross-functional teams in fast-paced, data-driven environments.

I am a quant developer and data engineer with deep experience building production systems for energy markets and large-scale analytics. I co-founded engineering efforts at Citadel Energy Marketing, owning volatility marking, pricing, scenario analysis, and valuation models for structured power and gas products.

Previously, I built streaming and batch data infrastructure at Pinterest, including the company's first Flink-based realtime experiment analytics pipeline and Spark-based data warehouse pipelines, and contributed to crowd-sourcing evaluation and machine learning research projects.

I blend strong quantitative skills from graduate training in statistics with hands-on software engineering across Python, Java, and Scala, and extensive experience with cloud and big-data stacks to deliver robust, scalable solutions that support trading, research, and analytics workflows.

Experience

Work history, roles, and key accomplishments

Current

Quant Developer

Current

Citadel

Jul 2021 - Present (4 years 2 months)

Tech lead for the volatility desk, built curve and volatility marking systems and valuation models for structured power and gas products, and developed a research library for market-data and scenario analysis supporting pre-trade to P&L attribution workflows.

Python MongoDB S3 Redis

Data Engineer

Citadel

Jun 2020 - Jun 2021 (1 year)

Built data pipelines, APIs and realtime monitors for commodities desks (Power, Natural Gas, FTR, Weather) using S3, MongoDB, MSSQL and realtime scraping tools to support trading workflows and analytics.

Python S3 MongoDB MSSQL Redis Selenium BeautifulSoup Pandas

Software Engineer

Jul 2018 - May 2020 (1 year 10 months)

Built Pinterest's first Flink-based streaming experiment analytics pipeline processing terabytes/day and developed the first data-warehouse pipeline and Hive-to-Spark migrations to improve ETL performance.

Apache Flink Kafka Spark Hive HBase SparkSQL Data Pipelines ETL

Software Engineer Intern

Jun 2017 - Sep 2017 (3 months)

Integrated F_beta into crowd accuracy tests and productionized two EM-based latent-variable aggregation models, increasing human-eval recall for minority classes by ≥30%.

Python Hive Jupyter Latent Variable Models EM Algorithm EM Algorithms

Machine Learning Research Intern

Here

Jun 2016 - Sep 2016 (3 months)

Researched global change detection: improved map-matching and designed KNN and CNN methods (Caffe) for pixel-wise classification to predict map geometry and road attributes, and built a Flask-based evaluation tool.