Machine Learning Engineer – Quantitative Research & AI Systems

Mondrian Alpha; New York (NY); On-site; USD 120,000 - 170,000; Full time

Job summary

Mondrian Alpha is seeking a Machine Learning Engineer to innovate within the financial markets, focusing on deploying ML models for trading predictions and risk management. In this role, you will work closely with quantitative researchers to refine models and build out robust infrastructure that performs under real-time market conditions.

The ideal candidate will have significant production experience with ML systems, proficiency in Python, and a strong understanding of model architecture and performance analysis. Exceptional compensation and an elite team await the right individual.

Qualifications

• Experience building and deploying machine learning systems in production environments.
• Proficiency in Python and core ML tools and frameworks.
• Understanding of modeling, assumptions, and limits.

Responsibilities

• Research and develop machine learning models for financial prediction and risk.
• Build production infrastructure reliably from research models.
• Collaborate with quantitative researchers to expedite learning from data.

Skills

Machine learning systems deployment

Python proficiency

Understanding of modeling and algorithms

Software engineering fundamentals

Tools

PyTorch

JAX

Job description

Most places that say they are serious about machine learning are serious about the word. This firm is serious about the work. For years, ML here has not been a strategy deck or a transformation initiative - it has been in production, making decisions, every single day, in one of the most adversarial and information-dense environments on earth: financial markets. The researchers and engineers here are not doing ML because it is fashionable. They are doing it because it works, and because they have built an environment where the rigor required to make it work is treated as a baseline, not an aspiration. If you have spent your career pushing the frontier of what models can do, this is where that instinct gets truly tested.

The Opportunity

We are looking for a Machine Learning Engineer who is done optimizing for scale metrics and ready to optimize for something harder - alpha. You will work at the intersection of cutting-edge ML research and real-world deployment, building and owning systems where model performance has an immediate, measurable, financial consequence. There is no proxy metric here that can be gamed. The market tells you if you are right. That feedback loop, brutal as it is, is also what makes this environment one of the most accelerating places a machine learning practitioner can spend their time.

What You'll Do

• Research, develop, and productionize machine learning models applied to financial prediction, signal generation, execution optimization, and risk
• Build the infrastructure that takes models from research to production reliably - training pipelines, feature stores, model serving, monitoring, and retraining workflows
• Work directly alongside quantitative researchers to understand what they are trying to learn from data, and build the systems that let them learn it faster and deploy it with confidence
• Push on model architecture, feature engineering, and experimental design with the same rigor you would apply at a top research organization
• Identify failure modes before they matter, instrument everything, and build systems robust enough to perform in regimes the training data never anticipated
• Help define what state-of-the-art ML infrastructure looks like in a production financial environment - this is a build, not a maintenance role

What We're Looking For

• Deep, hands-on experience building and deploying machine learning systems in demanding production environments - not prototypes, not demos, systems that run
• Strong proficiency in Python and the core ML stack; comfort with PyTorch, JAX, or equivalent frameworks at a research-engineering level
• Genuine understanding of modeling - not just the APIs, but the math, the failure modes, the assumptions, and the limits
• Experience bridging the gap between research and production - you have taken models from a notebook into the world and kept them working once they got there
• Strong software engineering fundamentals: you write code that other exceptional engineers are happy to read, review, and build on
• The intellectual honesty to know when a model is not working, the curiosity to understand why, and the persistence to fix it

Nice to Have

• Experience with time-series modeling, sequential prediction, or other domains where the data generating process is non-stationary and adversarial
• Familiarity with reinforcement learning, online learning, or adaptive systems that update in response to live feedback
• Background in NLP, large-scale representation learning, or generative modeling applied to structured data
• Prior experience in a quantitative finance, trading, or market-facing ML environment
• Experience with high-performance feature engineering at scale and low-latency model inference

Why This Role

The most intellectually honest thing we can say is this: if you want to know whether your models are actually good, come work somewhere that charges the market to find out. There is no benchmark you can overfit, no leaderboard that rewards the right tricks, no user study that can paper over a weak signal. There is only performance, measured in real time, against the most sophisticated adversaries in the world. For an ML engineer who has always wanted to know what they are really capable of, there is no better place to find out. Compensation is exceptional, the team is elite, and the work will demand - and develop - the best of