Want to connect with Flexpa?
Join organizations building the agentic web. Get introductions, share updates, and shape the future of .agent.
Is this your company?
Claim this profile to update your info, add products, and connect with the community.
Flexpa is an infrastructure provider that supplies the "ground truth" data necessary for AI agents to function in healthcare. They solve the primary bottleneck in medical AI: the inability for autonomous systems to access verified, structured patient histories. By providing a FHIR-first API with built-in consent and identity verification, Flexpa allows agent developers to programmatically ingest clinical records and insurance claims history.
In the agent stack, Flexpa sits in the data retrieval and context layer. They are active in championing the use of TEFCA and CMS mandates to empower agents in fields like clinical trial matching, benefits navigation, and legal subrogation. For anyone building a healthcare agent that needs to "know" a patient's medication list, previous diagnoses, or insurance coverage, Flexpa is the likely data source, moving the ecosystem from manual record retrieval to automated, API-driven workflows.
Flexpa is built on the premise that healthcare data is undergoing a fundamental shift from static, unstructured documents to real-time, machine-readable streams. This shift is driven by a series of federal mandates, specifically the CMS-9115-F and ONC (g)(10) rules, which require health insurers and providers to provide patients with digital access to their own data. Flexpa acts as the aggregator for these disparate endpoints, offering a single API that connects to what they call a "3-in-1 network."
This network covers three critical areas: health insurers (via the CMS mandate), medical record systems (via the ONC mandate), and nationwide data exchanges (via the Trusted Exchange Framework and Common Agreement, or TEFCA). Instead of a developer building hundreds of custom integrations for different hospital systems or insurance carriers, they connect to Flexpa. The platform handles the identity proofing—using IAL2 standards to verify who the patient is—and the consent management, ensuring that data is only shared when explicitly authorized.
The company began in 2020 as Automate Medical. Its original focus was more narrow, targeting the automation of pulmonary function tests (PFTs). The founders built tools like PFT Tutorbot and PFT Extractor to turn lung function reports, which were typically trapped in PDFs, into structured JSON data. This early work led them into the world of healthcare standards and the FHIR (Fast Healthcare Interoperability Resources) community. Realizing that the data interoperability problem was universal across healthcare, the team refocused on building broader developer tooling, eventually rebranding as Flexpa.
CEO Andrew Arruda previously co-founded Ross Intelligence, a legal AI startup that was a pioneer in using natural language processing for legal research. His background in highly regulated, document-heavy industries informs Flexpa's approach to healthcare. CTO Joshua Kelly, formerly of Universe (acquired by Live Nation), provides the infrastructure expertise needed to handle high-scale data transactions. The company is based in San Francisco and raised a $8.5 million seed round led by General Catalyst in 2022.
For builders in the AI agent space, Flexpa solves the "cold start" problem of patient context. An AI agent tasked with finding the right clinical trial or optimizing a Medicare Advantage plan is useless without the underlying clinical and claims history. Flexpa provides this context in a format—FHIR-based JSON—that is natively compatible with modern LLM-driven workflows.
Existing competitors like Particle Health or Zus Health also operate in the data aggregation space, but Flexpa differentiates by leaning heavily into the "patient-directed" model and the specific branding of "records for agents." They are betting that the next generation of healthcare services will not be driven by doctors clicking buttons in a portal, but by autonomous agents that require clean, high-velocity data feeds to make decisions on behalf of patients. Their customer base currently includes clinical research organizations like Curavit and digital health platforms like Evidation, where instant record retrieval replaces months of manual work.
A single API for patient-consented health records from insurers, EHRs, and exchanges.
This project provides the source for the SQL on FHIR v2.0 Implementation Guide
A Model Context Protocol implementation for FHIR
Benchmarking Large Language Models for FHIR
Medplum is a healthcare platform that helps you quickly develop high-quality compliant applications.
Public feedback and feature requests for Flexpa products
An example of using Flexpa Link in a React Native application
Flexpa is hiring
You've explored Flexpa.
Join organizations building the agentic web.