RabbitHawk Labs
The applied decision-science lab behind RabbitHawk
Research-backed, not research-trapped. Labs studies forecasting, optimization, agentic AI and human-controlled decision systems, and every method is judged by one test: does it improve decisions under real-world constraints?
- publications across the team and research network
- 1,100+publications across the team and research network
- citations
- 50,000+citations
- highly cited research
- Top 1%highly cited research
Lab mission
Turn research-grade methods into enterprise decision systems
The same team whose work is used at scale by leading organizations builds RabbitHawk. Labs is where forecasting science, optimization, applied AI and alignment research become tools planning teams actually trust, tested against decisions, not benchmarks alone.
Total verified output: 1,181 publications and 53,485 citations, detailed on the Research & Decision Science page.
Scientific principles
What the lab believes
Uncertainty should be visible
Plan for ranges, not false precision. Every forecast carries honest confidence.
Forecasts must reconcile
Local plans add up to executive truth across every hierarchy.
Optimization must respect constraints
Budgets, capacity, lead times and service levels are first-class inputs, not afterthoughts.
Context changes decisions
Signals outside structured systems move the answer, so the system reads them.
Humans stay in control
Recommendations are auditable; people approve and override high-impact decisions.
Every decision should teach the system
Outcomes feed back, so the next decision is sharper than the last.
Research themes
What we work on
Probabilistic & hierarchical forecasting
Uncertainty-aware forecasts that reconcile across SKU, store, region and time.
Constraint optimization & operations research
Decisions that respect real budgets, capacity and trade-offs.
Agentic & contextual AI
Reading unstructured signals and turning them into auditable forecast shifts.
Human-in-the-loop decision systems
Approval, override and explainability for high-impact decisions.
Alignment science
Getting teams to agree on goals worth optimizing for: the front end of the loop.
Explore the lab
Podcast, methods and notes
Have a hard technical problem?
Bring us a forecasting, optimization or applied-AI problem and we’ll discuss feasibility, approach and a path to production.
Discuss a hard problem