Machine Learning Scientist [Analyst/Associate]

About the job

As a Machine Learning Scientist on the AI team at Cerberus, you’ll work on high-impact projects that combine the pace of a startup with the reach of a global investment platform. Our team partners directly with internal investment desks as well as portfolio companies across industries to deliver machine learning solutions that unlock value and accelerate decision-making.

Your work will range from developing and validating robust predictive models for pricing and valuation across diverse asset classes to dynamically optimizing prices under changing market conditions. You’ll be expected to translate complex data into actionable insights and ensure your solutions are not only technically sound but also adopted and delivering measurable business value, supporting deal team members and portfolio company executives.

We’re looking for machine learning scientists who are passionate about impact—those who bring deep statistical knowledge, thrive in fast-paced environments, and want to see their models deployed, used, and making a difference.

What you will do

• Build and deliver AI solutions: Design and implement advanced models and systems as both an individual contributor and as part of cross-functional teams.
• Drive impact through execution: Apply a hypothesis-driven approach to design solutions, collaborate with technical teams, and deliver results that create measurable business value.
• Work in an agile, fast-paced environment: Rapidly iterate and adapt to changing priorities, using creativity and pragmatism to maximize outcomes.
• Leverage modern tools and methods: Develop innovative solutions using contemporary platforms, languages, and frameworks, and package IP into reusable components.
• Communicate insights effectively: Translate complex technical concepts into clear, compelling narratives that drive understanding and action across technical and non-technical audiences.
• Build trust through delivery: Establish credibility by delivering high-quality solutions, challenging assumptions constructively, and iterating quickly in response to feedback.
• Develop broad technical capability: Work across the full data science lifecycle, continuously learning and applying new technologies.

Sample project you will work on:

• Real estate portfolio valuation: Work on developing advanced valuation models for real estate portfolios using internal and external data sources. This includes building predictive models with uncertainty estimates, improving model performance through rigorous evaluation, and creating data pipelines to support modelling and analytics. You’ll also prototype processes downstream of valuation models, such as optimization approaches, to enhance pricing strategies, collaborating with stakeholders to integrate these solutions into business processes.
• Price optimization & forecasting for goods: Develop machine learning models to forecast demand and optimize pricing strategies for goods sold by a portfolio company. You’ll build predictive models that incorporate seasonality and competitive pricing data, while quantifying uncertainty and maintaining model explainability to support robust, transparent decision-making.

Your Experience:

We’re a small, high-impact team with a broad remit and diverse technical backgrounds. We don’t expect any single candidate to check every box below - if your experience overlaps strongly with what we do and you’re excited to apply your skills in a fast-moving, real-world environment, we’d love to hear from you.

• Strong technical foundation: Degree in a STEM field (or equivalent experience) with hands-on expertise in a least two of applied statistics, machine learning, forecasting, NLP, or optimization. Experience with uncertainty quantification, model evaluation, and statistical inference is highly valued.
• Python expertise: Skilled in building data pipelines and ML models using modern libraries across multiple domains:
• Data science stack: NumPy, pandas / polars, scikit-learn, XGBoost, LightGBM
• Deep learning: PyTorch, JAX
• Statistical programming: NumPyro, PyMC
• Data skills: Proficient in SQL, with the ability to write efficient, maintainable queries and manage data pipelines for analytics and modelling workflows.
• Model development & deployment: Familiarity with deploying models into production environments, collaborating with engineering teams, and using tools like MLflow or Weights & Biases for experiment tracking and reproducibility. Proof of work in cloud environments, especially MS Azure, is a plus.
• Research mindset with business impact: Ability to translate complex problems into tractable modelling approaches. Strong problem-solving skills, intellectual curiosity, and a pragmatic approach to delivering solutions that drive measurable business value.
• Collaboration and Communication: Demonstrated experience working in collaborative development environments using tools like Git and Azure DevOps. Comfortable contributing to shared codebases, participating in code reviews, and managing branches and CI/CD workflows. Proven ability to work cross-functionally with data scientists, engineers, and non-technical stakeholders to translate business needs into technical solutions and ensure successful delivery and adoption.

About Us:

We are a new, but growing team of AI specialists- data scientists, software engineers, and technology strategists - working to transform how an alternative investment firm with $65B in assets under management leverages technology and data. Our remit is broad, spanning investment operations, portfolio companies, and internal systems, giving the team the opportunity to shape the way the firm approaches analytics, automation, and decision-making.

We operate with the creativity and agility of a small team, tackling diverse, high-impact challenges across the firm. While we are embedded within a global investment platform, we maintain a collaborative, innovative culture where our AI talent can experiment, learn, and have real influence on business outcomes.