CrewAI Multifamily Demand Lead Qualification & Voice Scheduling Agent with Bland AI
Multifamily leasing teams lose qualified prospective tenants due to delayed phone inquiry responses during peak leasing windows. Build an agent system using CrewAI and Bland AI voice calls to handle tenant inquiry calls and update leasing pipeline states. Achieve 90% automated lead resolution within 30 seconds.
What you are building
The core problem, expected build, and operating context for this challenge.
Develop a multi-agent orchestration pipeline using CrewAI and Bland AI API to automate tenant inquiry qualification and tour scheduling.
How work is evaluated
Evaluates lead qualification accuracy and tour scheduling tool calls based on call transcript evaluations.
Challenge datasets
Public datasets, plus the private datasets you have access to.
How submissions are scored
These dimensions define what the evaluator checks and which criteria separate a passable run from a strong one.
lead_qualification_match
Validates qualified status outcome
This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.
resolution_rate
Percentage of valid tenant lead queries correctly processed • target: 0.9 • range: 0-1
This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.
What you should walk away with
Configure CrewAI agents with defined roles (Lead Qualifiers, Tour Schedulers, Leasing Knowledge Assistants)
Integrate Bland AI API to trigger dynamic voice calls for lead follow-ups and preference gathering
Parse unstructured voice call transcripts into structured guest card CRM payloads
Handle edge cases such as budget mismatch, move-in date conflicts, and specialized amenity requests
Reference links and supporting material
Simulated voice phone transcripts between prospective apartment tenants and AI leasing agents.
Challenge setup
Evaluates lead qualification accuracy and tour scheduling tool calls based on call transcript evaluations.
Challenge input
JSON containing caller transcript, requested move-in, and target budget
CrewAI
Agent orchestration framework with role-based execution.
Bland AI
Telephony voice interface for automated caller interactions.
Evaluated output
JSON containing qualified_status, tour_scheduled, and preferred_unit_type
- Validates qualified status outcome
- Percentage of valid tenant lead queries correctly processed • target: 0.9 • range: 0-1
- Benchmark: Multifamily Lead Automation Benchmark
- Resolution Rate target: 90%
- 1 public reference case
- Python execution harness
- Python sandbox (unavailable on Versalist)
Summary shows the first task, up to three tools, and up to three evidence items. Open the technical recipe to view all configured items.
View technical recipe
Configured tools
- Bland AI · Required
- CrewAI · Optional
- Aim · Optional
Evaluation contract
- lead_qualification_match · Weight 1
- resolution_rate · Weight 1
Metrics
- resolution_rate · Target 0.9
Benchmarks
- Multifamily Lead Automation Benchmark
- general
Reference cases
- 1 public reference case
Workflow tasks
- Task 1
Input: JSON containing caller transcript, requested move-in, and target budget
Output: JSON containing qualified_status, tour_scheduled, and preferred_unit_type
Execution harness: Python
Evaluation tools
- CrewAI · agent_framework
- Bland AI · voice_ai
Recipe freeze marker
No recipe freeze time is recorded.