AgentHire – AI-Powered Candidate Screening & Interview Orchestration
Inspiration
Recruitment teams often spend significant time manually reviewing resumes, comparing candidates against job requirements, identifying skill gaps, and preparing interview plans. This process is repetitive, time-consuming, and difficult to scale when handling large volumes of applications.
We wanted to demonstrate how agentic AI and BPMN orchestration can streamline hiring workflows while keeping decision-making transparent and structured. Our goal was to build an intelligent hiring assistant that automatically evaluates resumes, matches candidates to job descriptions, generates personalized interview plans, and routes candidates through a hiring workflow.
What it does
AgentHire is an AI-powered hiring workflow built on UiPath Maestro BPMN that automates candidate screening and interview planning.
The solution performs the following steps:
- Candidate uploads a resume.
- Resume Analyzer Agent extracts skills, evaluates projects and experience, generates a candidate score, and creates a professional candidate summary.
- Candidate Matcher Agent compares the candidate profile against a job description, identifies matched and missing skills, and calculates a match score.
- Interview Planner Agent generates interview rounds, technical questions, and a final recommendation.
- BPMN orchestration routes the candidate through appropriate hiring paths:
- Approved
- Rejected
- Manual Review
This creates a complete end-to-end recruitment assessment workflow.
How we built it
The solution was built using UiPath Maestro BPMN and multiple AI agents working together in a coordinated workflow.
Resume Analyzer Agent
- Extracts technical skills
- Evaluates projects and experience
- Generates candidate score
- Produces candidate summary
Candidate Matcher Agent
- Compares resume analysis with job requirements
- Identifies matching skills
- Identifies missing skills
- Calculates candidate-job match score
Interview Planner Agent
- Creates interview rounds
- Generates personalized technical questions
- Produces final interview recommendation
BPMN Workflow
The agents are orchestrated through UiPath Maestro BPMN, enabling structured process execution and automated decision routing.
Challenges we ran into
- Designing reliable data flow between multiple agents.
- Standardizing JSON outputs across agents.
- Mapping agent outputs correctly inside BPMN processes.
- Configuring gateway conditions for automated routing.
- Handling agent execution errors and workflow debugging.
- Ensuring consistent recommendations across all stages of the hiring pipeline.
Accomplishments that we're proud of
- Successfully built a multi-agent hiring workflow.
- Automated resume analysis and candidate evaluation.
- Generated personalized interview plans dynamically.
- Integrated AI agents with UiPath Maestro BPMN orchestration.
- Created a scalable recruitment process with clear decision paths.
What we learned
Throughout the project, we gained practical experience with:
- UiPath Maestro BPMN orchestration
- Multi-agent workflow design
- Agent-to-agent communication
- Process automation architecture
- Structured decision routing
- AI-driven hiring workflows
We also learned the importance of maintaining consistent data contracts between agents and designing workflows that can scale while remaining easy to understand.
What's next for AgentHire
Future improvements include:
- Automated interview scheduling
- AI-generated offer letters
- Candidate ranking dashboards
- Recruiter approval workflows
- ATS integration
- Email and notification automation
- Hiring analytics and reporting
- Candidate feedback generation
Our vision is to transform AgentHire into a complete AI-powered recruitment orchestration platform that reduces hiring effort while improving consistency and candidate experience.
Built With
- agenticai
- json
- machine-learning
- processorchestration
- recruitmentautomation
- uipathagentbuilder
- uipathautomathcloud
- uipathmasterobpmn
- workflowautomation
Log in or sign up for Devpost to join the conversation.