AI assistant for recruitment
AI assistant for recruitment,decisions made by people.
The AI recruitment assistant takes on the administrative side of selection: it summarises CVs against the job requirements, proposes interview slots and drafts messages to candidates. The recruiter reads, decides and is accountable for every step, and no rejection goes out without human review.
Recruiters get lost in admin while candidates wait
Every open role brings dozens or hundreds of applications, every candidate is waiting for a reply, and every interview takes several emails to find a time. The time for real conversations with candidates shrinks, communication is delayed, and good candidates accept other offers. At the same time, candidates' personal data requires careful handling.
- Reading every CV against the same requirements takes hours for each role
- Scheduling interviews needs many messages between candidates and managers
- Candidates do not receive timely updates on the status of their application
- Candidate data is scattered across emails and folders, making it hard to manage in line with GDPR
Who it is for
- HR teams that receive many applications for roles with clear requirements
- Companies that recruit continuously for operational roles, where interview scheduling takes a lot of time
- Recruitment agencies that want to respond to candidates faster without giving each one less attention
- Managers who recruit occasionally and need an orderly process, without a dedicated HR department
What the AI recruitment assistant does
- 01
Summarises CVs against the requirements
For each application, it shows which job requirements are explicitly covered and what information is missing, without a final verdict.
- 02
Orders applications for review
It proposes a reading order based on criteria strictly related to the role, and the recruiter sees every application.
- 03
Schedules interviews
It offers candidates slots from the interviewers' calendars and sends the invitations once confirmed.
- 04
Drafts messages to candidates
It drafts acknowledgements, invitations and status updates in the company's tone, approved by the recruiter.
- 05
Answers questions about the process
It explains the selection stages and the public details of the role to candidates, using approved information.
How it works
How the assistant supports a selection process
The steps marked in yellow are approved by a person.
- Input
- AI
- Human in the loop
- Action
- Result
- 01
Input
New application
The CV arrives through a job board, email or the careers page.
- 02
AI
Summary against requirements
The assistant notes which requirements are covered and what is missing.
- 03
Human in the loop
Recruiter decides
The recruiter reads the application and decides whether to invite the candidate.
- 04
Action
Interview scheduling
The assistant proposes free slots and sends the approved invitation.
- 05
Result
Candidate informed
The candidate receives a clear, timely message about the next step.
What information and tools the assistant uses
- The job description and the objective requirements set by the manager and HR
- The CVs and cover letters received, accessed only for the purpose of selection
- Interviewers' calendars, for proposing slots
- Approved message templates for each stage of the process
- The candidate privacy policy and data retention rules
Possible integrations
- eJobs
- BestJobs
- LinkedIn (within the limits of the available API)
- Workable
- Teamtailor
- Google Workspace (Gmail, Calendar)
- Microsoft 365 (Outlook, Teams)
- The careers page on your website
- eJobs
- BestJobs
- LinkedIn (within the limits of the available API)
- Workable
- Teamtailor
- Google Workspace (Gmail, Calendar)
- Microsoft 365 (Outlook, Teams)
- The careers page on your website
Each integration's compatibility is checked during the analysis phase, for the systems and versions you use.
Control and transparency
What stays in human hands
- Every decision to reject, invite to interview or hire, made by the recruiter or manager
- Reviewing all applications, including those placed lower in the proposed order
- Approving messages that communicate a selection outcome
- Setting the job requirements and checking that they include no discriminatory criteria
Limits and how we manage them
- A CV does not tell you everything about a candidate: experience described differently or atypical career paths can be underrated, which is why the recruiter sees every application
- The assistant does not use or infer criteria such as age, sex, ethnicity, marital status or photographs
- It does not make automated decisions that affect candidates; these would require additional safeguards under GDPR
- Using AI in recruitment may fall under strict rules in EU AI legislation, which are assessed together with the company's legal adviser
How we measure quality and fairness
We agree the indicators together before launch and track them afterwards. The values depend on the process, the data and the volume.
- Accuracy of summaries against the CV
- On a sample, recruiters check whether the requirements marked as covered actually appear in the CV.
- Consistency with the job requirements
- We compare the summaries for similar applications to identify unequal treatment.
- Checking for unequal effects
- We regularly analyse, on aggregated data and with HR's agreement, whether the proposed order disadvantages particular groups.
- Response time to candidates
- We measure the interval between applying and the first message, before and during the pilot.
- Messages edited by the recruiter
- We track how often and why drafts are corrected before sending.
How we implement the recruitment assistant
- 01
Process and risk analysis
We map the selection stages and the data used, and assess together with the data protection officer which safeguards are needed.
- 02
Requirements, templates and rules
We define the objective criteria for the role, the approved messages and the points where the recruiter decides.
- 03
Pilot on a single role
The assistant works in parallel with the recruiter, who compares the summaries with their own reading of the CVs.
- 04
Use and periodic audit
We extend to other roles and regularly check consistency, fairness and compliance with data retention.
Use cases
The scenarios below are conceptual examples, not delivered implementations.
- Retail chain
Recruitment for sales roles
The assistant acknowledges applications, summarises the stated availability and experience, and schedules interviews with store managers once they have decided.
Conceptual example - Automotive components factory
Selection for operators and technicians
For each CV, the assistant shows which of the required qualifications and certifications appear explicitly, and the HR team decides whom to invite.
Conceptual example - Software company
Communication with candidates
The assistant answers questions about the stages of the process and sends status updates approved by the recruiter.
Conceptual example
Frequently asked questions
Does the assistant reject candidates automatically?
No. The assistant prepares summaries and suggestions, but every decision that affects a candidate is made and reviewed by a person.
How is candidate data protected?
Data is used only for the selection process the candidate applied to, with access limited to the team involved, clear retention periods, and candidates informed about the processing and the use of an AI tool, in line with GDPR.
How do you avoid discrimination?
We work only with requirements related to the role, exclude sensitive data from the analysis and regularly check that the results are consistent. The recruiter sees every application, not just those at the top.
What does the assistant do when a CV is unclear or atypical?
It flags it for careful reading by the recruiter and notes what information is missing, rather than automatically placing it at the bottom.
Can it be used for any type of role?
It is most useful for roles with clear requirements and a high volume of applications. For leadership or highly specialised roles, its part is usually limited to scheduling and communication.
Do we need to tell candidates that we use an AI tool?
We recommend that you do. Candidates are informed about the data processing and the use of an AI tool in the process, and we agree the exact wording with the data protection officer and the company's legal adviser.
Can’t find the answer? Write to us and we’ll reply directly.
Want your recruiters to have more time for people?
Describe what your selection process looks like today and we will look at which administrative part an assistant could take on, with the decision staying with your team.
Related solutions
- AI assistant for your internal knowledge baseEmployees ask how something is done in the company and get the answer from internal procedures and policies, with a reference to the source document.
- AI phone bookingBookings, rescheduling and cancellations made by phone directly in your live calendar, with a confirmation sent to the customer.
- AI agent for document processingTakes in incoming documents, extracts the data, checks it against your rules and systems and sends it on, with a person handling the exceptions.