Skip to content
RabbitHawk

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.

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