Pharmacy automation products — hardware and an agent workforce on one loop.
OneDose ships two machines — an IP dispensing machine that releases patient-specific doses at the point of care, and a dose packing machine that produces barcoded patient-specific packaging — and an AI agent workforce covering intake, refills, prior authorization, inbound calls, adherence and post-discharge follow-up. They run on one shared patient context and one audit log, which is what makes them a single medication loop rather than three purchases.
The machines close the loop inside the building. The agents close the segment past the door that no machine can reach.
OneDose CLMM — closed-loop medication management, extended past discharge.
OneDose CLMM is the closed medication loop assembled as one system: the order structured on arrival, the dose packed and barcoded against the verified order, released at the point of care against the patient, scanned before administration, written back in real time — and then followed past the discharge door, where every other closed loop in the category stops.
PharmaEye is the context layer underneath it. It doesn’t learn from your patients. It remembers them.
Preparation and packaging
The OneDose dose packing machine packages patient-specific doses and prints a barcode encoding the order it was packed against, so the package itself carries the identity the administration check needs.
Dispensing to the point of care
The OneDose IP dispensing machine releases patient-specific doses at the point of care and records the release against the patient, the nurse and the time, so an override becomes a logged exception rather than an invisible one.
Administration at the bedside
OneDose supports barcode-verified administration against the packaged dose and writes the result back. The clinical override decision stays with the nurse; the record of it does not stay with the nurse.
Documentation and reconciliation
Administration events are written back in real time with an immutable audit trail. Where a facility runs its own eMAR, OneDose writes to it rather than replacing it.
After discharge
Nothing in the category confirms that a discharged patient collected, understood, started, tolerated or continued their medication. The agent workforce runs the structured contact on every discharge — not the high-risk decile — and escalates what needs judgement to a clinician with the conversation attached.
It does not perform clinical verification, and it does not make the clinical decision at an override. Judging whether an order is safe for a patient is the clinical gate of the whole medication process, and it stays with a pharmacist. Any vendor offering to automate that step is offering you a liability, and any accuracy figure presented as making that check unnecessary should end the conversation.
What are the machines?
Each has its own page describing how it works, what it records, and — the part worth reading before a demo — what it deliberately will not do.
OneDose IP Dispensing Machine
Patient-specific dose release at the ward, checked against the active order and recorded against the nurse, the patient and the time.
Inpatient dispensing hardwareOneDose Dose Packing Machine
Doses packed per patient per administration time, each pouch barcoded against the order it was packed from.
Packaging hardwareThe AI agent workforce
Six agents covering intake, refills, prior authorization, inbound calls, adherence and post-discharge follow-up — on the same patient context as the machines.
Software
Which product solves which problem?
These get sold interchangeably and they address different stages, so it is worth being blunt. The most common expensive mistake in this category is solving a well-understood problem with the adjacent product.
A preparation problem — hand-sorted rounds, unlabelled split tablets, doses that cannot be scanned because nothing barcoded them — is a packing problem. A cabinet does not fix it.
A release and attribution problem — drifting controlled-substance counts, doses that cannot be traced to a patient, overrides nobody reviews — is a dispensing problem. A packing machine will produce beautifully packaged doses and the same reconciliation failures.
An availability problem — a unit waiting on pharmacy for a time-critical dose — is a stocking problem, and the right answer is often a cabinet rather than anything on this page. We would rather say so than sell you a machine that makes your emergency department slower.
A visibility problem after discharge — no idea whether the patient collected or continued the medication — is not a hardware problem at all. It is the agents.
Frequently asked
- What products does OneDose make?
- OneDose ships two machines — the IP Dispensing Machine, which releases patient-specific doses at the point of care against the active order and records every release, and the Dose Packing Machine, which produces barcoded patient-specific packaging — together with an AI agent workforce covering intake, refills, prior authorization, voice, adherence and post-discharge follow-up. They run on one shared patient context and one audit log.
- Do the machines work without the AI agents?
- Yes. The dispensing and packing hardware closes the packaging, dispensing and administration stages of the medication loop on its own, and most deployments start with one machine on one unit. The agents add the segment past discharge, which is the part no machine can reach — so the two are worth more together than separately, and neither requires the other to function.
- Does OneDose replace our EHR or pharmacy system?
- No. OneDose reads and writes against the systems you already run rather than replacing them — the verified order comes from your pharmacy system and the administration record goes back to your eMAR. Integration depth varies by system, and OneDose does not publish a list of named integrations it has not confirmed per system. Ask and it will be confirmed in writing before you commit to anything.
- What do the machines cost?
- OneDose does not publish hardware pricing, because the honest figure depends on the number of care areas, which stages you already have, and what your EHR and pharmacy vendors charge for the interfaces. The costs that get underestimated in this category are consistently the interface work and the workflow redesign rather than the equipment, and any vendor quoting a per-bed number before seeing your dispensing footprint is quoting one they will revise.
- Can you publish throughput and capacity specifications?
- Not on this page, and deliberately. Throughput, canister capacity and cycle time are measured values that depend on product mix and changeover frequency, and a figure published without its measurement conditions is a marketing number that a pharmacy director can falsify with a stopwatch. Ask, and what exists will be shown with the conditions it was measured under and against a product mix resembling yours.