Real AI vs process automation

Automation replays your process. AI agents understand the order.

Process automation is a recording of what your team already does. When the situation changes -- a new VMS format, an unusual shift requirement, an edge-case compliance question -- it stops. Globus AI agents read context and decide, the way an experienced consultant would.

How they differ in practice

When the situation changes, one breaks and one adapts

Feature Globus AI agents Process automation
Reading a new VMS format or email layout Reads and understands the content Fails or requires reprogramming
Matching candidates Searches and reasons across context Applies fixed rules set at configuration
Explaining a recommendation Plain-language reasoning shown to the consultant No explanation -- output only
Compliance edge cases Flags and surfaces for human review Passes or fails on rule match
Adapting to new client requirements Updated in conversation with your team Requires developer changes
Human in control
Eira -- order intake

Understands any format, not just the ones it was trained on

Process automation reads the third cell in column B. Eira reads what the shift actually says -- the clinical grade, the ward context, the dates, the rate -- regardless of format, source, or phrasing. When the VMS portal updates its layout, Eira carries on. When a new client sends orders as voice notes via WhatsApp, Eira adapts.

  • Email, portal, PDF, VMS extract, message -- any source, any format
  • Captures clinical grade, ward, dates, rate, and compliance requirements
  • Handles ambiguity by flagging for consultant review, not failing silently
Eira
Eira Inbox Intelligence
Saga -- candidate matching

Reasons across your database, explains every result

A rules engine scores candidates on the fields you told it to score. Saga searches across CVs, booking history, notes, availability, and compliance status -- and shows you why each candidate is on the shortlist. When the order is complex or unusual, Saga surfaces the trade-offs rather than silently omitting candidates.

  • Full database search, not just a filtered list
  • Every shortlist position explained in plain language
  • Unusual requirements surfaced for consultant judgement
Saga
Saga Search & Match
Ingrid -- compliance

Tracks what matters, flags what is about to matter

Rules-based compliance checking tells you when something has already expired. Ingrid watches NMC PIN, DBS, right-to-work, mandatory training, and framework-specific requirements -- and alerts your team before expiry becomes a problem. She maintains a full audit trail and surfaces the relevant check for any client or framework requirement.

  • Proactive expiry alerts, not just pass/fail checks
  • NMC PIN, DBS, right-to-work, and mandatory training in one place
  • Full audit trail for framework and client reviews
Ingrid
Ingrid Compliance
  • GDPR compliant
  • UK & EU data storage
  • Human-in-the-loop design
  • EU AI Act aligned

Common questions

We already use automation tools. Does Globus replace them?

Not necessarily. Globus AI agents sit at the layer where context and judgement matter -- order intake, candidate matching, compliance edge cases. Deterministic automation for payroll, invoicing, and data sync can stay in place. Many agencies run both.

How long does it take to set up?

Agents connect to your existing ATS and can be configured without developer work. Your team works with our team to define each agent's scope and guardrails before they go live.

What does the walkthrough actually show?

We take a real shift order, run it through Eira and Saga live, and show you the reasoning at each step -- the match, the compliance check, the explanation. Two minutes. Your data if you have it, a realistic sample if not.

See the difference

2-min agent walkthrough -- live, no slides

We run a real shift order through Eira and Saga and show you what the agents see, decide, and explain. Book a time and we will walk you through it.