Can Oncology Workflows Run Without Human Touch?

on YouTube ↗17m

" summary

Claims end-to-end automation of oncology prior authorization using an LLM-driven stack: an ingestion agent, EV agent, medical necessity agent and a submission browser agent that orchestrate as a DAG and apply a deterministic decision engine plus confidence scoring to escalate only uncertain cases. Uses LLM-generated RPA configs, a payor rule SQL KB, a self-healing loop and 100+ browsers on Kubernetes to drive portals, handling thousands of submissions per hour and deployed across 20+ hospitals and 100,000 patients, with measurable no‑touch share growth.

AI-written summary · about

" chapters (8)

1. 0:00Automating oncology workflows
2. 1:24Eliminating human review
3. 2:51Unified insurance service
4. 6:00Drug authorization requirements
5. 11:34Medical necessity reasoning
6. 14:05No-touch order submission
7. 15:50Expanding workflows
8. 16:12Future of no-touch automation

" related talks

• 46mLeveraging AI and Automation in DevOpsYishai Beeri, Michael Kaufmann, Anurag Kale, Albert David Lewandowski · DevDays Europe 2024
• 25mLLM Powered Agents with KubernetesHema Veeradhi, Shrey Anand · Cloud Native + Kubernetes AI Day NA 2024
• 14mOrchestrating Agentic Workflows with Governance and TrustXin Ru Lee · SuperAI Singapore 2025
• 10mAI-Powered Dental Claims ProcessingPravesh Nikhare · Conf42 Large Language Models (LLMs) 2025
• 50mHow to Test LLM AgentsJames Massa · TestMu Conference 2025
• 19mEveryday AI: Leveraging LLMs for Simple...Matthew Sullivan, Dominic Zanardi · BSidesSF 2025
• 35mAgentic Workflows: Stories from the Engine RoomYvo Richner · webinale 2025
• 01h 56mCase Study + Deep Dive: Telemedicine Support Agents with LangGraph/MCPDan Mason · AI Engineer World's Fair 2025