Backed by Y Combinator.
Haladir is the operational AI layer for global logistics. We unify the data trapped across WMS, TMS, YMS, OMS, LMS, IMS, ERP, WOS, EDI, and WCS/WES systems, deploy event-driven agents with execution authority, and embed solver-grade optimization into the core of 3PL and distributor operations.
Haladir ingests, normalizes, and reconciles WMS, TMS, YMS, OMS, LMS, IMS, ERP, WOS, EDI, and WCS/WES systems into a single queryable operational graph. SKUs, orders, shipments, dock doors, and labor exposed to LLMs and solvers as first-class objects.
Event listeners on every flow surface the moments that matter within milliseconds. Policy-bounded AI agents release fulfillment waves, direct picking and packing, assign docks, and resolve exceptions autonomously inside guardrails operators define.
Solver-grade operations research and machine learning in production: vehicle routing (VRP, CVRP, VRPTW), multi-echelon inventory optimization (MEIO), mixed-integer programming (MILP), demand forecasting, ETA prediction, pick-path optimization, labor forecasting, and shift scheduling.
Primary market: 3PLs and distributors running live freight networks (global carriers, contract logistics, cold-chain, hazmat, and brand-owned fulfillment operators).
Secondary market: frontier AI labs licensing Haladir's substrate and engine as RL environments, post-training corpora, and evaluation harnesses for models that need to reason about the physical economy.