--- title: Can Oncology Workflows Run Without Human Touch? - Anant... description: An AI engineer at Trisca describes how they automated oncology prior authorization workflows end-to-end using a multi-agent system. The pipeline uses four... canonical: https://daily.dev/posts/can-oncology-workflows-run-without-human-touch---anant-shankhdhar-risa-labs-x1aqjicvu twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Can Oncology Workflows Run Without Human Touch? - Anant Shankhdhar, Risa Labs | daily.dev og:description: An AI engineer at Trisca describes how they automated oncology prior authorization workflows end-to-end using a multi-agent system. The pipeline uses four... og:url: https://daily.dev/posts/can-oncology-workflows-run-without-human-touch---anant-shankhdhar-risa-labs-x1aqjicvu og:image: https://api.daily.dev/og/posts/x1AqjicVU.png og:image:alt: Can Oncology Workflows Run Without Human Touch? - Anant Shankhdhar, Risa Labs og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Can Oncology Workflows Run Without Human Touch? - Anant Shankhdhar, Risa Labs **[AI Engineer](https://daily.dev/sources/aidotengineer)** · 16 min read · 0 upvotes · 0 comments ## Summary An AI engineer at Trisca describes how they automated oncology prior authorization workflows end-to-end using a multi-agent system. The pipeline uses four agents: an eligibility verification agent, an authorization status agent, a medical necessity agent, and a submission agent. Key techniques include a unified payer integration service with LLM-driven config generation and self-healing RPA, multi-source evidence reconciliation to build confidence scores for drug authorization decisions, a payer rule knowledge base built from portal checks and LLM extraction, and a medical necessity agent that queries a patient biomarker graph to answer clinical questions with supporting evidence. Orders that pass all checks with sufficient confidence are submitted without any human review, while uncertain cases are escalated to clinicians. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### How can LLM agents be used to automate prior authorization workflows in healthcare without human review? Four specialized agents handle the process: an eligibility verification agent checks insurance coverage, an auth agent determines whether a drug needs authorization, a medical necessity agent evaluates clinical justification against policy criteria and patient biomarkers, and a submission agent files the request. Deterministic checks and multi-source evidence reconciliation (authorization letters plus a payer rule knowledge base) allow confident cases to skip human review while low-confidence cases still escalate. _daily.dev surfaces engineering writeups like this for teams designing agentic automation in regulated domains._ ### Why can't LLM extraction alone eliminate human review in prior authorization automation? LLM extraction from patient notes is inherently non-deterministic, so its outputs cannot be blindly trusted and still require human verification even if efficiency improves. Reliability only increased once a second independent evidence source, such as prior authorization letters and a payer rule knowledge base built from portal checks, was added so agreeing signals could raise confidence enough to skip human review. _Developers weighing LLM reliability in automated pipelines can track patterns like this on daily.dev._ ## Similar posts on daily.dev - [How AI Can Improve Insurance Verification Workflows](https://daily.dev/posts/how-ai-can-improve-insurance-verification-workflows-fn4yt2gsx) · Software Testing Magazine · 2 upvotes · 0 comments - [AI automation for hospitals to save both time and cost](https://daily.dev/posts/ai-automation-for-hospitals-to-save-both-time-and-cost-pegonppkc) · Ruby Flow · 0 upvotes · 0 comments - ["OncoAgent: A Dual-Tier Multi-Agent Framework for Privacy-Preserving Oncology Clinical Decision Support"](https://daily.dev/posts/oncoagent-a-dual-tier-multi-agent-framework-for-privacy-preserving-oncology-clinical-decision-supp-1jnbxv9zb) · Hugging Face · 2 upvotes · 0 comments - [Rehumanizing global health care with agentic AI](https://daily.dev/posts/rehumanizing-global-health-care-with-agentic-ai-xzrofzo6q) · MIT Technology Review · 0 upvotes · 0 comments - [AI Workflow Automation for Dev: Hand an Agent a Whole Job](https://daily.dev/posts/ai-workflow-automation-for-dev-hand-an-agent-a-whole-job-iil79phay) · Telerik · 1 upvotes · 0 comments --- Tags: [#llm](https://daily.dev/tags/llm), [#ai-agents](https://daily.dev/tags/ai-agents), [#automation](https://daily.dev/tags/automation), [#robotics](https://daily.dev/tags/robotics), [#healthcare](https://daily.dev/tags/healthcare) [View this post on daily.dev](https://daily.dev/posts/can-oncology-workflows-run-without-human-touch---anant-shankhdhar-risa-labs-x1aqjicvu) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Can Oncology Workflows Run Without Human Touch? - Anant Shankhdhar, Risa Labs","url":"https://daily.dev/posts/can-oncology-workflows-run-without-human-touch---anant-shankhdhar-risa-labs-x1aqjicvu","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/can-oncology-workflows-run-without-human-touch---anant-shankhdhar-risa-labs-x1aqjicvu"},"datePublished":"2026-07-20T06:39:37.195Z","dateModified":"2026-09-14T07:27:05.599Z","description":"An AI engineer at Trisca describes how they automated oncology prior authorization workflows end-to-end using a multi-agent system. The pipeline uses four...","image":"https://i.ytimg.com/vi/_cVfz88_j7A/sddefault.jpg","thumbnailUrl":"https://i.ytimg.com/vi/_cVfz88_j7A/sddefault.jpg","isAccessibleForFree":true,"articleSection":"AI Engineer","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"AI Engineer","logo":"https://media.daily.dev/image/upload/s--u5PucxNT--/f_auto/v1724338940/logos/aidotengineer","url":"https://daily.dev/sources/aidotengineer"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/can-oncology-workflows-run-without-human-touch---anant-shankhdhar-risa-labs-x1aqjicvu","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"llm,ai-agents,automation,robotics,healthcare","timeRequired":"PT16M","video":{"@type":"VideoObject","name":"Can Oncology Workflows Run Without Human Touch? - Anant Shankhdhar, Risa Labs","description":"An AI engineer at Trisca describes how they automated oncology prior authorization workflows end-to-end using a multi-agent system. The pipeline uses four...","thumbnailUrl":"https://i.ytimg.com/vi/_cVfz88_j7A/sddefault.jpg","uploadDate":"2026-07-20T06:39:37.195Z","duration":"PT16M","url":"https://api.daily.dev/r/x1AqjicVU","embedUrl":"https://www.youtube.com/embed/_cVfz88_j7A"}} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"AI Engineer","item":"https://daily.dev/sources/aidotengineer"},{"@type":"ListItem","position":3,"name":"Can Oncology Workflows Run Without Human Touch? - Anant Shankhdhar, Risa Labs"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/can-oncology-workflows-run-without-human-touch---anant-shankhdhar-risa-labs-x1aqjicvu#faq","mainEntity":[{"@type":"Question","name":"How can LLM agents be used to automate prior authorization workflows in healthcare without human review?","acceptedAnswer":{"@type":"Answer","text":"Four specialized agents handle the process: an eligibility verification agent checks insurance coverage, an auth agent determines whether a drug needs authorization, a medical necessity agent evaluates clinical justification against policy criteria and patient biomarkers, and a submission agent files the request. Deterministic checks and multi-source evidence reconciliation (authorization letters plus a payer rule knowledge base) allow confident cases to skip human review while low-confidence cases still escalate. daily.dev surfaces engineering writeups like this for teams designing agentic automation in regulated domains."}},{"@type":"Question","name":"Why can't LLM extraction alone eliminate human review in prior authorization automation?","acceptedAnswer":{"@type":"Answer","text":"LLM extraction from patient notes is inherently non-deterministic, so its outputs cannot be blindly trusted and still require human verification even if efficiency improves. Reliability only increased once a second independent evidence source, such as prior authorization letters and a payer rule knowledge base built from portal checks, was added so agreeing signals could raise confidence enough to skip human review. Developers weighing LLM reliability in automated pipelines can track patterns like this on daily.dev."}}]} ```