krv labs

Privacy-first AI for clinical trials and research.
_{We build models that learn informative representations of patient data, and run them where that data lives, so researchers can find the right patients, design better trials and generate clinical knowledge without moving sensitive records.}

---

🧬 What We Do

• Foundation models: representations of the patient journey (labs, medications, diagnoses, encounters over time) that transfer across diseases, sites and studies.
• Decision models: calibrated, interpretable decisions built on those representations, such as who is eligible, which cohort to study and what is likely to happen next.
• Local intelligence: inference runs inside the hospital or on the researcher's machine, not in someone else's cloud.
• Privacy first: patient data never has to leave the institution to be useful.

Krv stands for Kepler, Riemann and von Neumann. Krv Labs is built to honor the lineage of giants who refused to accept "good enough," and we bring that standard to healthcare AI.

📁 About This Repository

This is the organization's shared .github repository:

• profile/README.md: the public profile shown at github.com/Krv-Labs
• .github/ISSUE_TEMPLATE/: default issue templates for all org repositories
• .github/workflows/: reusable GCP build, deploy and PR staging workflows
• templates/: starter Dockerfile and cloudbuild.yaml
• docs/: CI/CD setup guides (Workload Identity Federation, staging environments)

---

Backed by NVIDIA Inception, Berkeley SkyDeck, PAD-13, and TUM Venture Labs.