EV Charging and eMobility Software

OCPP 1.6J and 2.0.1. The specification is the easy part — the difficulty is that charger manufacturers implement it inconsistently, so a message that works against one vendor fails against another. Real OCPP work is as much about handling vendor-specific behaviour and building a test harness against actual hardware as it is about the protocol itself.

Yes, and that is usually the requirement. A CPO estate is rarely single-vendor. The approach is an abstraction layer over OCPP that normalises vendor differences, plus a certification process for each new charger model before it goes live. Skipping that step is why mixed estates produce intermittent faults nobody can reproduce.

OCPI for network interoperability, so your drivers can charge on other networks and theirs can charge on yours. It covers location and tariff publication, authorisation, and charge detail record exchange for settlement. The engineering is manageable; the commercial agreements and reconciliation disputes are usually the harder part, and the system has to make those disputes resolvable.

Usually. If the hardware speaks OCPP, we build against it and certify each model. Where a manufacturer uses a proprietary protocol, integration depends on what they document or expose — sometimes an API, sometimes very little. We establish that per model during discovery rather than assuming a mixed estate behaves uniformly.

Useful than exact. State of health models built on charge cycles, temperature history, charging rates and depth of discharge give a defensible estimate that improves as data accumulates. They will not match a laboratory capacity test. For fleet residual planning and warranty decisions that is generally sufficient, and it is considerably better than mileage as a proxy.

For a defined scope we target a working MVP in 30 days and a core production build in 90. EMobility extends when charger certification across multiple vendor models, payment licensing or grid connection constraints are involved. Hardware availability for testing is a common and underestimated cause of delay — we flag it before starting.