bayesian workflow
- prior elicitation & generative model design
- full bayesian model building (Stan, PyMC)
- posterior predictive checks & decision analysis
infrastructure for adaptive decision-making
services.
we build systems that learn from data and support better decisions under uncertainty. every engagement is scoped to the problem, not a template.
the practice of sam cialek
book a consultation →
the work
01–03
causal inference
- DAG specification & causal model design
- treatment effect estimation from observational data
- A/B test design & sensitivity analysis
emergent simulation
- agent-based model architecture
- synthetic population generation & calibration
- scenario simulation & policy testing
applied across domains
4 domains
sustainability & esg
emissions accounting (scope 1/2/3), LL97 & SB253 compliance, decarbonization planning, GRESB benchmarking, net-zero roadmaps
health technology
wearable + clinical data integration, causal health models, EHR middleware, personalized intervention design
political & social modeling
disposition modeling, preference geometry, synthetic population simulation, opinion dynamics
finance & risk
portfolio inference, risk quantification, alternative data pipelines, decision systems under uncertainty
deliverables
3 formats
models & pipelines
production-grade inference systems, reproducible workflows, documented codebases
reports & analysis
technical write-ups, executive summaries, investor-grade documentation
dashboards & tools
interactive visualizations, decision support applications, monitoring systems