AgentHack: qBotica - TalenTransformer - Multimodal Agents (Voice, Video)

AgentHack submission type

Enterprise Agents

Name

Russel Alfeche

Team name

qBotica TalenTransformer

Team members

@devik @akhil_padgilwar_qbotica @zell12

How many agents do you use

Multiple agents

Industry category in which use case would best fit in (Select up to 2 industries)

HR
Universities/Academy

Complexity level

Advanced

Summary (abstract)

We present a multi-agent multimodal HR recruitment automation pipeline designed to streamline and intelligently manage the candidate interview lifecycle—from identity verification, to interview proper, to compensation recommendation. This solution uses a modular set of UiPath agents to automate interview initiation, screening, scoring, dashboarding, and compensation analysis, enabling faster and more data-driven hiring decisions.

Detailed problem statement

Manual hiring processes are time-consuming, prone to bias, and inefficient at scale. Key challenges include:
• Verifying candidate authenticity during virtual interviews
• Ensuring fair and consistent evaluation across candidates
• Managing technical interviews at scale
• Tracking interview results in a structured, real-time manner
• Recommending compensation aligned with market benchmarks
Organizations lack a secure, scalable, and intelligent system to automate this full workflow—especially when interviews are remote or asynchronous.

Detailed solution

Our solution addresses this with an agentic approach, where each intelligent UiPath agent is responsible for a specific phase of the hiring workflow.
Process Flow:

  1. Interview Notification: Sends personalized interview invites with unique links. Comes from upstream agent that shortlists candidates and schedules the interview via voice enabled agent.
  2. Identity Verification Agent:
    o Name validation
    o Facial recognition for liveness detection
  3. Primary Interview Agent:
    o Initial HR Introductions and Questions
    o General Interview Questions and Answers
    o Coordination with/Passing to Technical Interview Agent
  4. Technical Interview Agent:
    o Technical and Domain-specific questions
    o Agent based/judgement scoring
  5. Queue Transmission:
    o Forwards results to Orchestrator Queue
  6. Insights Dashboard Agent:
    o Presents interview results on UiPath Insights
  7. Compensation Research Agent:
    o Recommends salary offers using market data
    The final output is a dashboard-driven candidate report, including pass/fail status, scores, and compensation offers.

Demo Video

Expected impact of this automation

• 80% reduction in manual HR screening time
• 100% candidate identity verification before interview
• Consistent, bias-free scoring using AI/NLP models
• Centralized decision-making via dashboard
• Market-aligned compensation offers for selected candidates
• Can be scaled to 100s of candidates with minimal effort

UiPath products used (select up to 4 items)

UiPath Agent Builder
UiPath Coded Agents
UiPath Insights
UiPath Maestro

Automation Applications

HRIS, Decision Analytics & Compensation Planning

Integration with external technologies

Voice and Video Coded Agents (Livekit WebRTC, Langchain, STT-LLM-TTS Pipeline [Deepgram, OpenAI, Cartesia]), Market Salary Datasets or APIs (e.g., Glassdoor data, internal CSVs) for Agent Builder Tools and/or Context Grounding

Agentic solution architecture (file size up to 4 MB)

Sample inputs and outputs for solution execution

Inputs:
• Interview ID (Main Reference to HR Entities in the Database)
• HR Interview/Candidate Database (HRIS)
o Candidate Table – Candidate profile or information including the name, total experience, current salary, expected salary, notice period, etc
o Interview – Specific instance of Interview associated with a Candidate and a Job position
o InterviewResults – Results summary of the conducted interview
o JobPositions – Contains all information about the open position including industry, job level, title, skills required, etc
o Technical & HR question sets (for interview agent’s context)
• Camera Feed for Identity Verification
• Salary datasets (CSV or API) for compensation research agent’s context
Outputs:
• Identity verification result
• Interview scores (Primary/Technical)
• Queue item with full interview metrics/scores/feedback summary
• UiPath Insights dashboard metrics
• Compensation recommendation (numeric + tier)

Other resources

Project Artifacts

6 Likes