Quantitative Researcher

The Role

• •We are looking for an exceptional researcher with a strong mathematics and causal modeling background and serious practical instincts. The core work is the design and rigorous validation of causal graphical model architecture: developing principled solutions to hard inference problems and evaluating them empirically against ground truth outcomes.
• •The standard is a first-rate PhD or MS from a leading institution in mathematics, computer science, or mathematical statistics. We are looking for someone who finds these problems genuinely compelling and attacks them with the rigour and creativity that characterises the best work in the field.
• •You will have significant freedom to define and pursue your own research directions. Good ideas get built. The pace is fast and the feedback loop from research to live system is short. If that environment energises you, this role is worth a conversation.

Responsibilities

• •Causal graphical model design. Develop and validate the architecture of probabilistic graphical models over structured event data. This includes model specification, identifiability analysis, and systematic evaluation of architectural choices against resolved outcomes.
• •Probabilistic inference. Research and implement inference methods appropriate to our model class. Work spans exact and approximate inference, uncertainty quantification, and the behaviour of estimators under short, noisy observation windows, and algorithm acceleration to support real-time processing.
• •Empirical validation. Design and own the evaluation framework, and capable of using cloud machines to access data and run repeatable experiments. Run structured ablation studies using proper scoring rules against ground truth data. Establish what each modelling choice contributes and maintain calibration discipline throughout.
• •Signal integration. Develop methods to combine heterogeneous probabilistic signals from multiple sources of varying quality and coverage. Characterise uncertainty honestly when signals are weak or conflicting.
• •Research to production. Translate validated research into production systems. Work closely with the engineering team to ensure model behaviour in production matches research expectations.

Requirements

• •First-rate PhD or MS in mathematics, computer science, or related from a leading research institution
• •Deep fluency in probabilistic modelling and inference: the kind that comes from years of serious mathematical work, not familiarity with standard toolkits
• •Demonstrated ability to formulate novel problems precisely and make meaningful independent progress on them
• •Strong empirical instincts: equally comfortable at the blackboard and with real, messy data
• •Rigorous approach to evaluation; hard to fool by spurious results
• •Proficient in Python and the scientific computing stack
• •Thrives with autonomy; sets their own agenda and drives to conclusions without needing external structure
• •Ships fast and often into production.

Advantageous

• •Familiarity with copula models, survival analysis, or causal graphical models.
• •Experience with probabilistic time series or embedding models.
• •Prior work in any domain where getting the probability wrong had measurable real-world consequences.

What We Offer

• •Research autonomy: define your own directions, own your work end to end, see it ship to production
• •Problems at the frontier of causal inference and probabilistic AI, grounded in one of the most incentive-aligned and data-rich information sources available
• •Intellectual depth without bureaucracy: a small team where mathematical rigour is the standard and good ideas move fast
• •Founding research hire with meaningful equity
• •Fast-moving culture where the feedback loop from idea to live system is measured in days, not quarters
• •Flexible working model

Apply

Send your resume or CV and a short note to [email protected].