T-NEX
Menu
AI agents for companies

Pre-sort applications, answer questions, support onboarding

T-NEX builds AI agents for HR work: screening and pre-sorting applications, answering staff questions on leave, policies and procedures and providing onboarding material. The agents access only approved HR data.

Discuss your HR use caseExplore the details
AI agents for HR, recruitment and learning platforms
  1. 01What the agents take on in HR work
  2. 02Only approved HR data, only for the intended audience
  3. 03Components for applications and staff questions

Aligned with your processes

In daily work

What the agents take on in HR work

The agent works with what your HR department already holds: incoming applications, policies, leave rules and onboarding material. Which of these areas we start with is decided in the project.

  • Screen applications: read the documents and transfer the details needed for the role into fixed fields, such as qualifications, previous positions and availability.
  • Pre-sort applications: against criteria your HR team sets for each role, the agent produces a proposal that points back to the original. Unclear cases go to a person for review, and the decision stays with the team.
  • Answer questions about leave, for example how a request is made and who signs it off, from the current policy and with a source reference.
  • Answer questions about policies and procedures, such as the travel policy or ordering work equipment, with a reference to the policy itself.
  • Provide onboarding material: new staff find checklists, contacts and access procedures for their first weeks, without HR answering every question one by one.
  • Pass questions without a sourced answer to the responsible person in the HR team instead of guessing.
Overview

Only approved HR data, only for the intended audience

HR data is sensitive. Before anything is connected we therefore define which sources the agent may use and which staff may see what. An answer must not reveal content that is not intended for the person asking.

For applications, the pre-sorting is a proposal that your HR team confirms or corrects. Missing and unclear details are made visible rather than silently filled in. Which details from applications an AI model may receive, and whether personal data is needed for that at all, is clarified in advance and recorded in the project documents. The same applies to log contents, retention and the question of who gets access for set-up and support.

Before operation, a pilot tests the agent on ordinary questions, alternative wording and questions without a matching answer, and for applications on difficult documents as well. Your HR team assesses whether answers are correct, whether the source supports them and how much human intervention is actually needed.

Overview

Components for applications and staff questions

For applications we use Document AI. It recognises the document type and the required details and presents unclear passages for review. A review interface should bring the extracted details together with the original. Each document type gets an agreed target schema, and difficult cases such as poorly legible passages or missing values are included from the start.

For staff questions the AI chatbot with source references is used. It finds maintained answers and shows their stored source location. A supplementary RAG function can compose answers from document passages; that stage is developed and tested as a separate scope.

Behind it, AI knowledge management sets up the upkeep of knowledge pages on policies and onboarding material as an ongoing process. Content owners are assigned to each source, and a change to a policy can trigger a fresh review of linked knowledge pages and test questions.

Overview

More providers for this task in Malaysia

The register of AI providers in Malaysia lists further providers for this task, including solutions for HR administration, payroll, recruitment and learning platforms. This page describes only what T-NEX builds itself.

All 14 tasks for which T-NEX develops AI agents are in the overview.

The workflow

How we introduce the solution.

  1. 01

    Choose a use case: We begin with a defined area, such as the most frequent questions to the HR team or the applications for one type of role, and check whether reliable material and responsible content owners exist.

  2. 02

    Set data access: Approved sources, audiences, the details a model may receive, log contents and retention are fixed before any HR data is connected. We also clarify in advance which data may be used in testing.

  3. 03

    Pilot with a measurable goal: Together with your HR team we define what a correct answer or pre-sorting is and which errors are not acceptable. We measure quality, how well sources support the answers, correction effort and the human intervention needed.

  4. 04

    Operation and expansion: Only after the pilot do we extend the source collection or the audience. Questions without an answer expose gaps in the knowledge base and have a recipient, and new or changed policies go through a release step.

FAQ

Questions about the solution

Does the agent decide who gets hired?

No. The pre-sorting is a proposal based on criteria your HR team sets for each role. Unclear and missing details are made visible, and the team confirms or corrects the proposal against the original before it is used further. The decision on an application stays with the team.

Which HR data can the agent see?

Only the sources approved for the audience concerned. We clarify the access roles before sources are connected. If a source is missing, the agent says so, asks a clarifying question or hands over to the responsible person.

Can the agent look up in the HR system how much leave a person has left?

That requires a connection to the system concerned. Additional system actions are planned separately, and the field mapping is defined beforehand. A chat window alone does not create such a connection. Without one, the agent answers from the approved documents, such as the leave policy.

Related options

You may also be interested in these.

Which task would you like to solve next?

Bring a concrete task. Together, we will define what the application needs to do.

Discuss your HR use case