Machine Learning Researcher

Thurn Partners; Miami (FL); On-site; Full time

Job summary

A quantitative trading firm is seeking a Machine Learning Researcher to join their systematic equities research team. You will design and implement training and evaluation pipelines for large language models while exploring alignment and reinforcement learning techniques. The ideal candidate has a strong background in machine learning and programming, though prior finance experience is not required. This role emphasizes both research quality and real-world performance, providing a unique opportunity in a research-driven environment.

Qualifications

• Strong background in machine learning, optimization or reinforcement learning.
• Deep understanding of transformer architectures and attention mechanisms.
• Experience working with post training techniques such as SFT, DPO or RLHF.
• Comfort working with large datasets and complex experimental pipelines.

Responsibilities

• Design and implement large scale training and evaluation pipelines for language models.
• Explore alignment methods such as RLHF and DPO.
• Improve model reasoning and accuracy in noisy datasets.
• Optimize models for GPU efficiency and precision.

Skills

Machine learning

Optimization

Reinforcement learning

Python programming

Deep learning frameworks (PyTorch, TensorFlow)

Analytical thinking

Job description

A highly sophisticated quantitative trading firm is seeking a Machine Learning Researcher to join a systematic equities research team focused on applying large scale language models to real world financial decision making.

The organization operates at the intersection of advanced machine learning and global markets, investing heavily in frontier AI research to develop models capable of reasoning over complex, noisy and dynamic datasets. Researchers in this group work closely with quantitative scientists and engineers to design systems that translate cutting edge research into measurable trading impact.

This role offers the opportunity to work on some of the most challenging problems in applied machine learning while operating in an environment where research quality and real world performance carry equal weight.

The Role

• You will design and implement large scale training and evaluation pipelines for language models, spanning pre training, domain adaptation and post training optimization.
• The work will involve exploring alignment and reinforcement learning techniques to adapt models for domain specific objectives, as well as investigating ways to improve reasoning, reliability and factual consistency in complex environments.
• Researchers in this group also experiment with retrieval systems and embeddings to connect models with external knowledge sources and proprietary datasets. Alongside research work, you will collaborate with engineering teams to ensure that successful approaches can be deployed into production research workflows.

Key responsibilities include

• Designing training and evaluation pipelines for large language models
• Exploring alignment methods such as RLHF, DPO and related post training techniques
• Improving model reasoning, robustness and factual accuracy in noisy datasets
• Working with embeddings and retrieval systems to integrate external knowledge
• Optimizing models for practical constraints including GPU efficiency, precision types and parallelization
• Partnering with engineers and quantitative researchers to transition research into production systems

Experience

• Strong background in machine learning, optimization or reinforcement learning
• Deep understanding of transformer architectures and attention mechanisms
• Experience working with post training techniques such as SFT, DPO or RLHF
• Strong Python programming and experience with frameworks such as PyTorch or TensorFlow
• Comfort working with large datasets and complex experimental pipelines
• Strong analytical thinking and ability to translate research ideas into working prototypes
• Experience in finance is not required. Intellectual curiosity about financial markets and decision systems is valued.

The Environment

• The team operates within a research driven trading organization where engineers and scientists play a central role in generating competitive advantage. Significant investment is made in computing infrastructure, data acquisition and experimentation.
• Researchers are given the freedom to explore new approaches while working alongside experienced quantitative professionals focused on deploying ideas that translate into real world performance.
• Compensation and progression reflect the impact of your work and the technical depth you bring to the organization.