Trellis AI: Building AI Agents to Streamline Healthcare Paperwork

Trellis AI is tackling one of healthcare's most persistent problems: the mountain of paperwork that delays patient care. The Y Combinator-backed startup has developed AI agents that automate document intake, prior authorizations, and appeals processing for healthcare providers and pharmaceutical companies.

The company, founded in 2024 as a spinout from Stanford AI Lab, processes billions of dollars worth of therapies annually across all fifty states. Their AI agents classify medical referrals, understand chart notes, and automate contract and reimbursement searches to provide patients with accurate coverage determinations and cost responsibility information.

The Problem: Administrative Burden in Healthcare

Administrative costs account for over 20% of U.S. healthcare spending, creating delays in care delivery, revenue drain, and staff burnout. Healthcare providers have less visibility into patient care than ever before while drowning in paperwork. Prior authorizations alone can take days or weeks, delaying critical treatments for patients who need them.

Trellis aims to eliminate this pre-service paperwork burden by building what they describe as "the infrastructure behind how patients get medications." Their AI agent is trained on millions of clinical data points and converts messy, unstructured documents into clean, structured data that integrates directly with electronic health record (EHR) systems.

How It Works

Trellis's AI agents handle three core functions:

Document Intake Automation: The system processes various document types, extracting relevant information and structuring it for downstream workflows.

Prior Authorization Processing: By understanding medical necessity criteria and coverage policies, the AI can prepare and submit prior authorization requests automatically.

Appeal Management: When authorizations are denied, the system can identify appealable cases and generate appropriate documentation for resubmission.

Market Traction and Growth

The company reports having achieved 10x revenue growth in recent months and capturing a significant market share in specialty healthcare markets. They serve patients across all fifty states and are scaling to hundreds of healthcare locations.

Trellis has attracted backing from prominent investors including Y Combinator, General Catalyst, and Telesoft Partners, along with executives from Google and Salesforce. The founding team includes alumni from Stanford AI Lab with experience ranging from international physics olympiads to founding engineering roles at unicorn startups.

The Human Impact

Beyond the technical achievement, Trellis emphasizes the real-world impact of their work. Healthcare operations leaders on their team have overseen 50+ healthcare locations, bringing deep domain expertise to the development of their AI systems. The company highlights that employees will "directly see the number of patients who received treatment because of the agents you built."

This focus on measurable outcomes reflects a broader trend in healthcare AI: moving beyond experimental pilots to production systems that handle critical healthcare decisions with robust evaluation frameworks.

Looking Forward

As healthcare providers continue to face staffing shortages and rising administrative costs, solutions like Trellis's AI agents could become increasingly essential. The company's growth trajectory suggests strong market demand for automation tools that can reduce the administrative burden while maintaining or improving accuracy in patient care workflows.

For healthcare organizations struggling with paperwork backlogs and prior authorization delays, Trellis represents a potential path to faster treatment times and reduced operational costs. The technology builds on advances in natural language processing and document understanding that have matured significantly in recent years, making large-scale automation of healthcare paperwork finally feasible.