Data Scientist (Bayesian Inference)

Numerator is seeking a Data Scientist (Bayesian Modeling) to help build, enhance, and scale data science services across our rapidly evolving data platform. You’ll work on initiatives that turn massive proprietary datasets into impactful, production-grade solutions..

This is a growth-track, product-focused role. You’ll collaborate with Product, Data, and Engineering teams to learn how customer needs translate into data-driven products, analytics methodologies, and new offerings that drive measurable business impact.

How You'll Spend Your Time:

• Contribute to the implementation and delivery of Bayesian and probabilistic modeling pipelines, from methodology research through production, with guidance from senior team members

• Execute on individual tickets independently and take on small epics with mentorship and guidance

• Work closely with Product, GTM, Data, and Engineering to turn models into reliable, production-grade solutions the business can depend on

• Actively participate in the team's learning culture (journal club, analysis reviews, standups) and seek feedback to continually level up your craft in Bayesian methods and reasoning about uncertainty

• Communicate methods, results, and tradeoffs clearly to both technical and non-technical audiences

• Strong foundation in Bayesian inference and probabilistic modeling — e.g. hierarchical / multilevel models, state-space and time-series models, graphical models, MCMC/HMC, variational and other approximate inference

• Experience or coursework applying probabilistic/Bayesian methods to real-world datasets, with a strong curiosity to learn production-grade standards

• Comfort reasoning about uncertainty, calibration, and model validation

• Facility with large or structured datasets and the computational side of inference at scale

• Strong Python, and fluency in a modern probabilistic-programming and numerical-computing stack — NumPyro, PyMC, Stan, JAX, dynamax, or similar. We hire on the ideas, not on exact tooling

• Demonstrated interest in shipping statistical models into production systems and writing maintainable code

• BS to PhD in Statistics, Math, Economics, Physics, CS, or a related quantitative field

• 0–2 years of relevant experience or recent graduate with strong quantitative project work

• Clear communication with both technical and non-technical audiences

Nice to Haves:

• Diagnosing and debugging large Bayesian models — convergence and divergence issues, pinning down which part of a big model is misbehaving, and knowing which inference method to reach for

• Weighting a non-representative survey or panel sample up to a known population, and a feel for where those adjustments break down

• Hierarchical models spanning multiple crossed or overlapping groupings — relationships that bridge hierarchies, not just a single nested tree

• Experience with graph or network models, or modeling relational / graph-structured data

• Measurement-error modeling, or reconciling multiple imperfect data sources

• CPG / FMCG / retail experience, or work with user-level purchase or panel data

What We Offer:

• An inclusive and collaborative company culture - we work in an open, transparent environment to get things done and adapt to the changing needs as they come

• An opportunity to have an impact in a technologically data-driven company that’s changing the market research industry and getting rave reviews

• Ownership of data solutions

• Market-competitive total compensation package

• Volunteer time off and charitable donation matching

• Strong support for career growth, including mentorship programs, leadership training, access to conferences and employee resources groups

• Regular hackathons to build your own projects and Engineering and Data Science Lunch and Learns

• Great benefits package including health/vision/dental, unlimited PTO, flexible schedule, internally quiet focus time, recharge days, 401K matching, travel reimbursement, and more