OneAI - Pipeline success - weekly meeting
Date: 2025-12-08 | Duration: 44.52000045776367 min | Account: pipeline-success
External attendees: bruna.santos@homecare.com, chris.b@pipelinesuccess.com, david@pipelinesuccess.com, ecorrons@homecare.com, jason.o@pipelinesuccess.com, ncuesta@homecare.com
Summary
The team discussed improvements to the inbound AI agent’s caregiver support flow, emphasizing clearer onboarding status inquiries and enhancing response accuracy to minimize confusion and misrouting. Chris Branum confirmed that the revised questions aim to clarify caregiver onboarding, while Amit Ben introduced a combination of scripted instructions and generative messaging to maintain a balance between guidance and conversational flexibility. The agent’s educational responses will be expanded, providing consistent information for both onboarded and non-onboarded caregivers. A deliberate slow rollout strategy was established to ensure quality and stability, with plans for full operation by mid-December. Additionally, the meeting addressed issues with duplicate API calls during transfers, necessitating urgent fixes. Participants agreed to increase meeting frequency to improve responsiveness and iteratively develop the agent’s capabilities.
Key Points
🤖 AI Agent Flow Improvements: Focus on onboarding questions and response accuracy to reduce confusion and misrouting for caregivers. 🚀 Controlled Rollout Strategy: AI agent deployment is slow for quality assurance, aiming for full operational status by mid-December. 📱 Text Messaging Limitations: Outbound texts are sent from a dedicated number, but two-way messaging is not yet supported. ⚠️ API Issues Identified: Duplicate API calls during transfers risk revenue; urgent fixes needed for consistent handling. 📅 Increased Meeting Frequency: Team will hold two to three additional meetings weekly to enhance responsiveness and accelerate development. 🔮 Long-Term Vision: Expand AI agent’s role for family support and integrate AI-driven quality control for better operational efficiency.
Overview
- AI Agent Flow Improvements: Focus on onboarding questions and response accuracy to reduce confusion and misrouting for caregivers.
- Controlled Rollout Strategy: AI agent deployment is slow for quality assurance, aiming for full operational status by mid-December.
- Text Messaging Limitations: Outbound texts are sent from a dedicated number, but two-way messaging is not yet supported.
- API Issues Identified: Duplicate API calls during transfers risk revenue; urgent fixes needed for consistent handling.
- Increased Meeting Frequency: Team will hold two to three additional meetings weekly to enhance responsiveness and accelerate development.
- Long-Term Vision: Expand AI agent’s role for family support and integrate AI-driven quality control for better operational efficiency.
Action Items
Chris Branum Lead testing and adjustment of inbound AI agent scripts incorporating new onboarding flow and education messaging (22:45) Move changes from production AI agent to test agent environment for clean testing (37:30) Provide updated onboarding and staffing page links to Amit for AI script inclusion (26:46)
Amit Ben Implement increased meeting cadence for collaboration (04:30) Adjust AI identity prompt and messaging instructions as discussed to improve flow and reduce confusion (10:10) Update AI script to include education for both onboarded and non-onboarded caregivers with text link and transfer option (22:45) Monitor calls and AI performance, prepare for regular review sessions on Wednesdays (36:05) Develop and propose AI call recording evaluation tool for quality control in transfer scenarios (42:00)
Daniella Block Schedule new midweek (Wednesday) recurring meeting slots for collaboration (36:05)
Jason O’Neill Coordinate participation of Noel on Wednesday meeting to address duplicate API calls related to transfer cases and sales errors (39:35) Continue overseeing phased rollout strategy prioritizing caregiver funnel completion mid-December and family onboarding thereafter (32:50)