AI agent · Human Talent

Smart Interviews

See how job interviews is evolving in organizations with AI agents.

  • Prebuilt agent Ready to use, we do not start from scratch.We start from an agent prepared for this process and configure it with your rules, your systems and your way of working.
  • Best practices Built in from day one.The agent's instructions incorporate the process practices we have applied when automating it in different organizations.
  • Configured by experts Functional and technology consultants.Specialized consultants support the configuration, integration, testing and go-live of the agent in your operation.
  • Agent platform Workforce AI.The agent runs on Workforce AI, GIITIC's own platform that centralizes security, permissions, audit, usage and administration of all your agents.
  • Process experience Processes + technology + enterprise software.Behind every agent are years of experience implementing enterprise software and automating processes in companies across Latin America.
  • GIITIC software ecosystem The tools the agent needs.We integrate with your current systems and, if you are missing one, the GIITIC ecosystem provides the modules the agent needs to query and act.
The current challenge

How it is done today, and why it is no longer enough

The structured interview is the selection procedure that best predicts job performance (r = 0.42, Sackett and others, 2022), but at many companies interviews are still free-form conversations that change depending on who conducts them.

  • Impossible schedules
  • Inconsistent interviews
  • Hours of first interviews
  • Incomplete notes
  • Questions left unexplored
  • Candidates who slip away
Meet your agent

Smart Interviewer agent

Smart Interviewer agent, an artificial intelligence agent

I'm an artificial intelligence agent trained by experts, ready to use and built with best practices. I am here to interview each candidate at the time they choose, by voice or chat, with the structured questions for the position: I listen, probe into their experience and competencies, answer their questions about the role and hand you their answers organized by criterion. You decide who moves forward.

Prebuilt, but adaptable to each organization.

How I work

  1. I detectCandidates readyI identify the candidates who reached the interview stage and invite them to choose their time.
  2. I understandPosition guideI read the profile, the criteria and the structured questions your team approved.
  3. I analyzeInterviewI talk with the candidate by voice or chat and probe deeper when their answer is vague.
  4. I decideAnswers by criterionI organize the information from each answer according to the defined evaluation criteria.
  5. I executeRecord and summaryI save the transcript, answers by criterion and a summary in your selection system.
  6. I verifyHandoff to the teamI notify the person responsible, flag incomplete answers and escalate to you the cases I couldn’t finish.
Results

Results you can achieve

more job offers with voice agent interviews
+12%Experiment with 70,884 applications for customer service roles: 9.73% received an offer versus 8.70%; people made the decisions.Source: Jabarian and Henkel, “Voice AI in Firms,” 2025
more candidates start working
+18%Applicants interviewed by the AI agent who start in the role, compared with those interviewed by people.Source: Jabarian and Henkel, “Voice AI in Firms,” 2025
higher retention at one month
+17%Probability of still being employed after at least one month when the interview was conducted by the AI agent.Source: Jabarian and Henkel, “Voice AI in Firms,” 2025
of applicants prefer the AI agent
78%Share who chose to be interviewed by the agent when they could choose between the agent and a person.Source: Jabarian and Henkel, “Voice AI in Firms,” 2025

Reference values based on benchmarks and comparable scenarios. Results depend on the characteristics of each operation and are not guaranteed results.

Greater speed, higher quality, lower costs

  • Greater speed

    Automate tasks and connect your information so your team responds sooner and your processes move forward without waiting.

    • Interviews each candidate at the time they choose, including outside office hours
    • Handles many candidates at once without matching schedules
    • Delivers organized answers as soon as the interview ends
    • Shortens the time between application and decision
  • Higher quality

    Standardize the work and control every step so your information is reliable and your results are consistent.

    • Asks the same structured questions to every candidate for the position
    • Probes for concrete examples when an answer is vague
    • Records the transcript and answers by criterion, with no scattered notes
    • Answers candidates’ questions with the approved information about the role
  • Lower costs

    Reduce rework, errors and manual work so your operation delivers more with the same resources.

    • Reduces the team’s hours spent on repetitive first interviews
    • Avoids losing candidates to scheduling delays
    • Handles application peaks without additional interviewers
    • Frees interviewers for the final conversations that add the most value
Before and now

Artificial intelligence in job interviews

Before

  1. 10:00 a.m.10:00 a.m. · MilestoneThe candidate is ready for the first interview
  2. 10:00 a.m.10:00 a.m. · Lost timeQueue of candidates to interview4 h
  3. 2:00 p.m.2:00 p.m. · PersonMessages the candidate to agree on a time10 min
  4. 2:10 p.m.2:10 p.m. · Lost timeWaits for the candidate’s reply and an open slot on the calendar20 h
  5. 10:10 a.m. · day 210:10 a.m. · day 2 · PersonConducts the phone interview30 min
  6. 10:40 a.m.10:40 a.m. · PersonTransfers their notes to the evaluation form15 min
  7. 10:55 a.m.10:55 a.m. · Lost timeThe notes wait between meetings and the manager’s other priorities6 h
  8. 4:55 p.m.4:55 p.m. · PersonThe manager reviews the notes and decides whether the candidate moves forward10 min
  9. 5:05 p.m. · day 25:05 p.m. · day 2 · MilestoneInterview recorded and decision made

Now

  1. 10:00 a.m.10:00 a.m. · MilestoneThe candidate is ready for the first interview
  2. 10:00 a.m.10:00 a.m. · AgentInvites the candidate to interview now or at a time they choose< 1 min
  3. 10:01 a.m.10:01 a.m. · Lost timeThe candidate chooses to start when their class ends1 h
  4. 11:01 a.m.11:01 a.m. · AgentInterviews the candidate by voice with the structured guide and probes into their experience25 min
  5. 11:26 a.m.11:26 a.m. · AgentRecords answers by criterion, transcript and summary< 1 min
  6. 11:27 a.m.11:27 a.m. · Lost timeThe manager finishes their current task1 h
  7. 12:27 p.m.12:27 p.m. · PersonReviews the agent’s summary and decides whether the candidate moves forward10 min
  8. 12:37 p.m.12:37 p.m. · MilestoneInterview recorded and decision made the same day

Tangible results in your operation

IndicatorBeforeNowFactor
End-to-end cycle time≈ 31 h≈ 2.6 h×12
Human work time65 min10 min×6.5
Waiting time30 h2 h×15
Human intervention100%≈ 22%÷4.5
Team capacity1×≈ 4.5××4.5

Illustrative reference scenario. Times depend on each operation and are validated with your data.

How work evolves

Your team moves from operating to deciding

The capacity the agent frees up goes to analysis, decisions, relationships, exceptions, improvement and innovation.

  1. No more low-value manual tasks

    • Matching schedules for first interviews through messages and calls
    • Repeating the same opening questions to every candidate
    • Taking handwritten notes and transferring them to the evaluation form
    • Chasing candidates who don’t reply
    • Interviewing outside working hours so as not to lose candidates
  2. The team adds value to the operation

    • Using final interviews to go deeper with the best candidates
    • Deciding with answers that can be compared by criterion
    • Improving the interview guide with what the answers show
    • Looking after the candidate experience throughout the process
    • Attending to candidates who prefer to talk with a person
  3. Your team's professional growth

    • Expert interviewerConducts the final interviews with the context the agent prepares.
    • Agent supervisorReviews the quality of the interviews, adjusts the questions and governs the agent’s knowledge.
    • Assessment designerImproves interview guides and criteria using hiring outcomes.
Enterprise knowledge

Agents are only as good as the knowledge and context they handle

It is one of the great challenges of enterprise artificial intelligence (AI): an agent without clear rules, reliable data and business context makes mistakes or stops. That is why knowledge is prepared, governed and improved as part of the project.

  • Structured interview guides by position
  • Competency evaluation criteria and scales
  • Approved information about the position, the company and the terms

The knowledge this agent handles

  • Answers to candidates’ frequently asked questions
  • Escalation rules and rules for treating candidates
  • Data protection and consent policies

60%

of AI projects without AI-ready data will be abandoned through 2026, according to Gartner. Gartner, “Lack of AI-Ready Data Puts AI Projects at Risk,” February 2025

Workforce AI helps you prepare, govern and improve it

  • Instructions prebuilt by experts
  • Process best practices
  • Tools to create and capture knowledge
  • Knowledge management
  • Integration with your current sources
  • Enrichment of existing knowledge
  • Ongoing updates
  • Knowledge governance
  • Continuous learning and improvement
Integration

We integrate with your environment

We connect the agent with virtually any application, API, service or data source in your company, whether GIITIC or third-party systems.

Integrations this agent requires

  • Applicant tracking or HR systemYour applicant tracking system (ATS), your HR system or GIITIC: candidates, stages, guides and the interview record.
  • Voice, chat and messagingCalls, WhatsApp, web chat or video calls where the candidate talks with the agent.
  • Calendar and emailInvitations, reminders and notices to the candidate and the hiring team.
Artificial intelligence (AI) chip

Your agent, connected to your operation

Base systems for agents

If you do not have the base system, we provide reduced versions of GIITIC modules with the capabilities the agent needs to work.

  • Applications and ERP
  • APIs and web services
  • Databases and files
  • MCP (Model Context Protocol), when applicable

An agent needs software tools to query information and carry out actions. Without them, it stays a conversation. That is why we cover both paths: connecting to what you already have or giving you what you are missing.

See GIITIC integrations

95%

of organizations struggle to integrate AI into their existing processes. MuleSoft (Salesforce), Connectivity Benchmark Report 2025, study sponsored with Vanson Bourne and Deloitte Digital

Workforce AI · Central governance platform

Today it is one agent, tomorrow dozens

Every area will add agents working at the same time. Workforce AI is the control center of your digital workforce, not an isolated script.

Explore Workforce AI
  • Agent creation and management
  • Security, identities and permissions
  • Policies and human oversight
  • Traceability and audit
  • Observability, quality and results
  • Models, tokens, usage and costs
  • Continuous updates to new AI models and features
Map of the Americas with the United States, Mexico, Guatemala, El Salvador, Costa Rica, Panama, Dominican Republic, Colombia, Venezuela, Ecuador, Peru, Bolivia, Chile and Argentina connected from Colombia
Our track record

Global experience at your service

Years of experience
15+
Companies in Latin America
850+
Countries
10+
Integrations
100+
ERP and ERP+ modules
40
AI agents
20+

Companies that trust GIITIC

Frequently asked questions

Questions about Smart Interviews

What interviews can the artificial intelligence agent conduct?
The first interviews for each position, by voice, chat or video call, with the structured guide your team approves: questions about experience, competencies, availability and terms. Channels and types of positions are confirmed when the project is scoped.
Does the agent decide who moves on to the next stage?
No. The agent interviews and hands you the answers organized by criterion, the transcript and a summary. A person always makes the decision; the consolidated evaluation of candidates can be supported by another Workforce AI agent.
What happens if the candidate prefers to talk with a person or there is a technical problem?
The candidate can ask for an interview with a person, and the agent coordinates it with your team. If the interview is interrupted, the agent resumes it or escalates it with what it managed to record.
Do I need GIITIC or can I use my current systems?
You can use your systems. The agent integrates with your applicant tracking or HR system, your channels and your calendars through APIs, services, files or MCP where applicable. If you don’t have the base system, we provide agent modules with the capabilities it needs.
How does it adapt to my positions and criteria?
The agent is prebuilt, but our consultants configure it with your interview guides, criteria, position information and rules for treating candidates, and test it with your cases before putting it to work.
Who controls what the agent does?
Workforce AI centralizes permissions by agent, human approvals, traceability of conversations and decisions, statuses and errors, auditing and usage control. You define what it asks, what it answers and what it escalates.
How long does implementation take, and how much does it cost?
It depends on your channels, systems, candidate volume and positions. Starting from a prebuilt agent avoids developing from scratch. First we talk and scope the project with you; then we prepare the proposal. We don’t publish prices.
What happens to my interviewers?
They stop repeating first interviews and matching schedules, and spend their time on final interviews and on deciding with better information. The agent expands their capacity; it doesn’t leave them without control.
Next step

Ready to see your agent work in your process?

Let's talk about your operation. A consultant will review your process, your systems and your knowledge, and show you how the agent would work with your rules.

  • Demo of the agent with cases from your process.
  • Review of your systems and the available knowledge.
  • Estimate of the scope and the next step.

Let's talk about your process

Fill in your details and a consultant will get in touch with you.

Fields marked with * are required.

Open in a new tab