Data Scientist

ABOUT THE ROLE

Calibrax AI builds production-grade AI systems that move the needle for our clients, from credit risk models in financial services to fraud detection, recommendation engines, and generative AI solutions. Our founding team consists of seasoned AI experts with years of hands-on experience and the bar for technical excellence here is genuinely high.

We are looking for a Data Scientist who builds things that work in the real world, not just in notebooks. You will own end-to-end model development across client engagements, from data exploration and feature engineering through to model deployment and ongoing optimisation. You'll work on a variety of high-stakes AI applications, often simultaneously, in environments where rigour and speed both matter.

WHO YOU ARE

You are a practitioner who takes pride in craft. You don't just run models, you understand why they work, where they break, and how to make them better. You're as comfortable discussing business metrics as you are optimizing hyperparameters, and you know the difference between a model that scores well in validation and one that actually delivers in production.

You think in systems. From data pipelines to deployment, you care about reproducibility, maintainability, and the full lifecycle of an ML product. You're curious about emerging techniques, especially in generative AI and LLMs, but grounded enough to apply them with judgment rather than hype.

You communicate clearly. You can walk a client's data team through your methodology or explain model outputs to a non-technical stakeholder with equal confidence.

MUST HAVES

• 3+ years of hands-on data science experience, with a strong portfolio of production ML models

• Deep proficiency in Python and the ML stack (scikit-learn, XGBoost, PyTorch, or TensorFlow)

• Experience in at least one of: credit risk / underwriting, fraud detection, NLP, recommendation systems, or computer vision

• Strong grasp of statistical fundamentals, you understand what your model is actually doing

• Experience deploying models in cloud environments (AWS, GCP, or Azure) and with MLOps practices

• Ability to work directly with clients and translate messy real-world data problems into clean modelling approaches

• Experience with LLMs, RAG pipelines, or generative AI applications is a strong advantage

WHAT SUCCESS LOOKS LIKE

• Your models ship to production and demonstrate measurable business impact for clients

• You produce work that is reproducible, well-documented, and maintainable by teammates

• Clients trust you as the technical voice in the room, you can explain your models and their limitations clearly

• You consistently identify the right modelling approach for the problem, not the most fashionable one

• You raise data quality issues early and help clients understand what 'good data' actually means for their use case

• You stay current with the field and bring relevant new techniques to the team, with appropriate context on when to apply them