Quantitative Developer – Compute & ML Platform

Job Description

Quantitative Developer – Compute & ML Platform | Millennium

The Tone:
This is a Quantitative Developer role at Millennium, a global, diversified alternative investment firm established in 1989. The company’s mission is to deliver results for investors through continuous evolution, innovation, and focus. This position is vital for the Execution Services team, contributing to the Central Liquidity Strategies group by building the foundational compute, storage, and model infrastructure. This infrastructure is essential for supporting advanced quantitative research and live trading across the platform.

The TL;DR
• Role: Full Time
• Team: Central Liquidity Strategies team within Execution Services
• Mission: To build the compute, storage, and model infrastructure that supports quantitative research and live trading for internal liquidity solutions.
• Tech Stack: Python, Ray, Dask, Spark, PySpark, Slurm, PyTorch, XGBoost, Git, Unix/Linux, Bash, CI/CD, kdb+, PyKX, Polars, Arrow, Kubernetes

What You’ll Actually Do
• API Extension: Extend the proprietary distributed dataset API, which features a Polars-style, lazily evaluated, DAG-planned DataFrame surface, with new operators, optimizations, and an improved new-user experience.
• ML Pipeline Implementation: Design and implement comprehensive machine learning pipelines covering large-scale training and tuning, model serving, production validation, and the full retraining lifecycle for a growing fleet of models.
• Infrastructure Improvement: Enhance the reliability, observability, and multi-user capabilities of the compute infrastructure, including resource management, workload isolation, and clean backend interfaces.
• Abstraction Building: Build durable abstractions that ensure the platform remains flexible and adaptable as distributed compute backends evolve over time.
• Engineering Practices: Write, support, maintain, and test code using strong engineering practices, encompassing unit testing, thorough documentation, automation, and CI/CD workflows.

The Must-Haves
• Background: Bachelor’s degree in Computer Science, Mathematics, Financial Engineering, Operations Research, or a related field, with a senior career level demonstrated by required experience.
• Experience: 6+ years of Python and distributed systems experience in a quantitative finance environment, with a strong track record of building platform abstractions used by others in production. This includes hands-on experience implementing machine learning and deep learning workflows using tools such as PyTorch and XGBoost, with comfort across the full training-to-serving lifecycle.
• Skills: Strong Python skills, deep working knowledge of one or more distributed compute frameworks such as Ray, Dask, Spark, PySpark, or Slurm, and strong working knowledge of Git, Unix/Linux, Bash, and modern CI/CD workflows, alongside strong communication skills for effective collaboration.
• Bonus: Experience with kdb+, PyKX, Polars, Arrow, columnar or lazy query engines, cash equities, live analytics, and/or cloud tooling including Kubernetes.