Minicor

What it does

Minicor is a platform for building and running desktop automations at scale using computer use agents. It enables AI companies to read and write to legacy desktop applications that have no APIs, such as EHRs, ERPs, DMS, and PMS. The platform provides self-healing agents that adapt to UI changes, built-in observability with video replays, and deployment options on Windows VMs, in the browser, on-premise, cloud, or Citrix. A single API call triggers a full desktop workflow and returns structured JSON.

Who it is for

Minicor is designed for AI companies selling into healthcare, automotive, logistics, and financial services that need to integrate with customers' legacy desktop systems of record. It is also for any organization that relies on old desktop applications with no writable API and needs to automate workflows reliably at scale.

Why it matters

Legacy desktop systems like EHRs, ERPs, DMS, and PMS often have no writable API, and some vendors actively restrict third-party API access. The only way to read or write data is to interact with the desktop like a human would. Building and maintaining RPAs in-house becomes unsustainable at scale because UIs change, edge cases pile up, and errors cascade. Minicor solves this by providing self-healing agents that achieve 93-96% click accuracy (vs. 80-85% for other approaches) and can go from zero to production in weeks instead of months.

Launch signal

Minicor launched on Hacker News as a Y Combinator-backed startup (YC P26). The website states it is SOC 2 Type II certified and HIPAA compliant. The platform is already in production with customers, processing 25,000 patients per day according to the site.

Brand and naming

The name "Minicor" suggests miniaturized or minimal core, aligning with the product's promise of lightweight, efficient automation. The tagline "Windows desktop automations at scale" clearly communicates the target platform and value proposition. The brand positions itself as a reliable, enterprise-grade alternative to traditional RPA, emphasizing self-healing and observability.