Reported operating use · Staffing and recruitment
AI recruiting shortlists: Adecco example
The Adecco Group reports a deployed recruiting workflow on Salesforce Agentforce that creates job orders from customer interactions, prepares interviews, summarizes calls and updates candidate records. Recruiters review shortlists. Reported results cover the broader agentic architecture, not a clean test of autonomous agents.
Job intake, candidate-record updates and recruiter shortlists
The Adecco Group
Deployed recruiting workflow, reported April 30, 2026.
A published third-party example, not a Nuvepro customer case or evidence of a partnership.
The workflow at a glance
The Adecco Group reported work
Illustrative summary of the published account
Reported work components. Their arrangement does not imply a fixed execution order or simultaneous work.
- Recruiting intake
Job orders and call summaries
The workflow uses customer interactions to prepare recruiting records.
- Record work
Candidate updates and interview preparation
Existing Salesforce data supports integration.
- Human handoff
Recruiter shortlists
Recruiters review proposed shortlists.
What the source reports
The published workflow creates job orders from customer interactions, prepares interviews, summarizes calls and updates candidate records. Recruiters review shortlists. Existing Salesforce data and workflows supported integration. Reported results cover the broader agentic architecture, not a clean autonomous-agent experiment.
Reported technology: Salesforce Agentforce. Scope and outcomes are source-reported.
What the enterprise environment would need
The following is a proposed project design for an enterprise doing similar work. Published details are marked; the remaining systems, data and scenarios would be agreed for your project.
Proposed practice environment
Prepare a job order and supported candidate shortlist
Illustrative project design to scope with your enterprise.
- 1 · Set up the environment
Recruiting intake bundle
Fictional call transcripts, job orders, candidate records and consent state.
- 2 · Preparation and agent contributions
Prepare order and shortlist
Extract requirements, reconcile contradictions and draft linked candidate updates.
- 3 · Independent checks
Check source annotations
Independent recruiter labels verify mandatory criteria, identifiers and lawful selection.
- 4 · People decide acceptance
Recruiter accepts version
Review requirements and consent before accepting the practice shortlist and updates.
A reviewable evidence package
A reviewed job order, draft candidate change set and shortlist rationale.
If a check fails → withhold acceptance.
A mandatory qualification is omitted or an impermissible criterion affects ranking. Withhold acceptance, repair, independently recheck and return to owner review.
Failure or owner rejection returns to review
- Hold
- → 2 · Revise work
- → 3 · Rerun independent checks
- → 4 · Owner review
Operating boundary. Recruiting test tenant only; no live candidate edits, outreach or selection decisions.
- Systems and tools
- Inferred test scope: Recruiting CRM test tenant with call transcript and email intake.
- Data and state
- Inferred test scope: Candidate profile, job order and shortlist share stable identifiers.
- People and handoffs
- Inferred test scope: Recruiter reviews proposed matches and corrects extracted requirements.
- Rules and authority
- Inferred test scope: Consent, access controls and lawful selection criteria require explicit review.
- Normal and exception cases
- Proposed test cases: Ambiguous pay range, contradictory qualifications, withdrawn candidate, duplicate job request.
- Reset and replay
- Not reported. Proposed: Seed synthetic candidate/job records; replay transcript versions without outreach.
A proposed simulation
An illustrative practice design derived from this work, not an executed assessment.
Convert a client call into a job order, reconcile contradictory requirements and update a candidate record while a recruiter reviews the shortlist.
Review gate: Check extracted requirements, identifiers, consent, lawful selection criteria and complete reviewer handoff; faster hiring is not match quality.
Challenge the result: Omit a mandatory requirement or apply an impermissible selection criterion. A convincing shortlist must not conceal the failure.
What needs checking
- Proposed: compare extracted fields with recruiter-reviewed source annotations.
- Proposed: inspect unauthorized record writes and shortlist explanations separately from speed.
- Proposed: recruiting owner checks representative intake language, exceptions and review practice.
Review environment fidelity, human capability and agent performance separately. The enterprise’s accountable owners decide acceptance of the shared workflow and any operating use.
Questions about this example
- What does Adecco use AI for in recruiting?
- The published workflow creates job orders from customer interactions, prepares interviews, summarizes calls and updates candidate records, using Salesforce Agentforce. Existing Salesforce data and workflows supported the integration.
- Where do recruiters stay involved at Adecco?
- Recruiters review shortlists. The source does not describe other review steps in detail. As a proposed check, not something Adecco reported, a recruiting owner would check representative intake language, exceptions and review practice.
- How strong is the evidence for Adecco's agentic hiring account?
- It is a deployed workflow reported on April 30, 2026. The reported results cover the broader agentic architecture, so they cannot be credited to agents alone. Independent selection quality, fairness and agent-specific causal attribution are not established.
- What would an enterprise need to build something similar?
- This is Nuvepro's proposed design, not Adecco's: a recruiting CRM test tenant seeded with synthetic candidate and job records, replaying transcript versions without outreach. Cases include ambiguous pay ranges, contradictory qualifications, withdrawn candidates and duplicate requests, checked against recruiter annotations.
Shape this around your enterprise
Bring a similar workflow and the people responsible for it. Together, define the records and systems to use, the result your team will build or configure, and how its owners will check it. Nuvepro scopes the simulation, environment and preparation. Agree any existing-agent evaluation, implementation assistance and production responsibilities separately.
Sources and limitations
Curation review: . This is the date this source example was reviewed, separate from source publication dates.
- Independent selection quality, fairness and agent-specific causal attribution are not established.
- Agentic hiring processes help the Adecco Group deliver faster placementsSource publication: 2026-04-30
More hr and recruiting examples
- AI candidate sourcing: Blue Castle Agency exampleBlue Castle Agency
- AI candidate sourcing: Anthropic exampleAnthropic
- AI hiring outreach: Alex Lieberman exampleHiring-outreach workshop example (Alex Lieberman)
- AI HR service desk: IBM exampleIBM