Causal Inference Scientist

About us: Ayass Bioscience is a CLIA-certified precision medicine company building AI platforms that turn transcriptomic data into causal, clinically actionable insight. BiRAGAS is our transcriptomic causal-inference research platform, applying rigorous causal methodology to genomic and biological data across multiple disease areas.

The role: You'll join the team building the causal inference core of BiRAGAS, moving analyses from association to defensible causation in gene-regulatory and disease contexts, working directly with our founder, our molecular biology team, and our engineering team.

What you'll do: Design, implement, and validate causal discovery and effect-estimation methods on bulk and single-cell transcriptomic data Contribute to the computational causal-modeling components of the platform, including graph-based and structural causal model approaches Design analyses that can withstand scientific and regulatory scrutiny, including appropriate robustness and sensitivity checks Develop benchmarks and evaluation criteria that quantify the strength of a causal claim and flag where evidence is insufficient or unsupported Translate model outputs into testable hypotheses for wet-lab validation and clinical collaborators Collaborate closely with our bioinformatics and data science teams, and contribute to publications, technical documentation, and partner-facing scientific materials

What you bring: Ph. D. in statistics, biostatistics, computational biology, computer science, or a related quantitative field (or equivalent experience) Deep working knowledge of causal inference: structural causal models, DAGs, do-calculus, causal discovery algorithms (PC, GES, NOTEARS, or similar), and identification of effects under confounding Hands-on experience with transcriptomic data (RNA-seq, scRNA-seq) and differential expression workflows Strong Python skills; comfort with PyTorch or JAX, and causal libraries such as DoWhy, causal-learn, or Causal Nex Ability to explain causal assumptions and evidence clearly to biologists, clinicians, and non-specialists A rigorous, detail-oriented approach to scientific reasoning

Nice to have: Experience with CRISPR perturbation data (e.g., CRISPR screens, Perturb-seq) or gene regulatory network inference Background in graph neural networks or knowledge graphs Background in immunology, oncology, or other complex/chronic disease biology Prior work building or validating evidence-grading frameworks Shape the causal backbone of a platform with over a decade of translational and clinical grounding behind it Direct access to a CLIA-certified lab for validating what your models predict Small, senior team where your methods ship into real research pipelines