Guests

• Claire Smid — AI Engineer, Xelix


• Emilija Gransaull — Back-End Tech Lead, Xelix


• Talal A. — Product Manager, Xelix

Key Takeaways

• Start narrow to win: pick high-volume, high-cost requests (invoice status & reminders).


• Enrichment > magic: accurate replies come from great retrieval/matching, not just a bigger LLM.


• Design for adoption: familiar inbox view helps onboarding, but a ticket-first UI unlocks AI features.


• Measure outcomes, not vibes: track “messages sent from Helpdesk”, % auto-resolved.


• Confidence builds trust: show match quality and response confidence so humans know when to edit.

0:00  Meet the Team: Claire, Emilija, and Talal
00:36  Introduction to Xelix and Its Products
01:08  Understanding Accounts Payable Teams
01:37  Help Desk Product Overview
03:11  Challenges Faced by Accounts Payable Teams
04:03  AI Integration in Help Desk
05:47  Automating Reconciliation Requests
07:45  Development Methodology: Carpaccio
09:11  Prototyping and Beta Testing
12:00  Manual Tagging and Data Collection
16:39  Focusing on High-Impact Use Cases
18:55  User Experience and Interface Design
24:56  Pipeline Architecture and Email Processing
28:21  Data Enrichment Challenges
29:04  Handling Vendor Identification
33:33  Email Thread Management
36:15  Generating Accurate Responses
40:48  Evaluating System Performance
49:20  Future Developments and Goals