Senior Biostatistician

Senior Biostatistician

We're hiring a senior biostatistician to lead our statistical modeling and predictive analytics work. This role owns the modeling architecture: designing and fitting longitudinal mixed-effects models, extracting coefficients and variance-covariance matrices, building dose-response curves, and powering the prediction engines behind our client-facing dashboards. You need to be able to do this work from day one. We're not looking for someone to train up — we need someone who has built these models before and can hit the ground running. This is a fully remote position.

Core Responsibilities

• Design, specify, and fit mixed-effects models (linear and generalized) for longitudinal observational data
• Extract beta coefficients and variance-covariance matrices from fitted models and export them for use in downstream prediction engines (JSON format → JavaScript calculators)
• Build and embed core calculator logic — the math that powers client-facing predictive dashboards
• Construct dose-response curves across product types, formulations, and outcome measures
• Handle model convergence issues, optimizer tuning, random effects specification, and model selection independently
• Cross-cohort and cross-product comparative analyses (global models, interaction effects, brand-level vs. cohort-level comparisons)
• Develop novel analytical approaches for new study types and research questions

• Build cohort-level industry reports — aggregate models across multiple products and brands within a study cohort
• Design cross-cohort comparative methodology as the dataset grows
• Cannabinoid predictability analysis: formulation × product type × dose → effect prediction
• Statistical architecture for new outcome domains (pain, oncology, non-cannabinoid substances)

• Own statistical methodology decisions for the company
• Validate and review analytical outputs
• Collaborate with the dev team on model-to-dashboard pipeline (you provide the statistical outputs, they build the software)
• Mentor the data analyst on advanced statistical concepts
• Document methodology for reproducibility and regulatory defensibility

Required Skills

• R (advanced, primary language) — must be fluent. This is not a Python shop.
• Mixed-effects models — lme4 package (lmer, glmer). Must understand random effects specification, convergence diagnostics, optimizer tuning (bobyqa, nelder-mead, nAGQ), model comparison, and cross-optimizer validation (allFit()).
• Coefficient extraction — ability to extract fixed-effect coefficients, variance-covariance matrices, and random-effect parameters from fitted models and structure them for use outside R (JSON, CSV, matrices).
• Longitudinal data analysis — repeated measures, within-subject designs, time-varying covariates, handling attrition and missing data
• Generalized linear mixed models (GLMMs) — binomial, Poisson, ordinal outcomes
• Dose-response modeling — experience building dose-response or exposure-response curves
• Statistical thinking at a senior level — ability to design analyses from scratch for novel research questions, select appropriate models, justify decisions, and interpret results in context. Not just running code someone else wrote.

Work Environment

• Fully remote
• Full-time (160 hrs/month)
• Reports to COO. Works alongside a full-time data analyst.
• AI-assisted workflow — we use AI tools extensively for coding and analysis. You should be comfortable leveraging AI tools and collaborating with AI-powered statistical review.
• Small team, startup pace — significant autonomy and direct impact on the company's analytical capabilities