What challenges did Every Cure face before AI implementation?

Identify new uses for existing drugs to treat roughly 14,000 diseases that lack FDA-approved therapies. Traditional repurposing studies are too slow and inefficient because they only examine a narrow subset of drug-disease pairs.

What was the impact of the AI solution?

Compressed a core prediction pipeline from about 100 days to about 17 hours. Scaled experimentation from three runs per year to dozens of experiments per month. Launched nine active drug repurposing programs within one year. Enabled scientists, physicians, and PhDs to review the top 6,000 predictions over the past year.

Source

GCPScaled productionEvidence: Medium65/100

Every Cure: Drug repurposing using Gemini Enterprise Agent Platform with BigQuery/Cloud Storage accelerated from 100 days to ~17 hours

Use case typeMultimodal analyticsUpdated Jul 1, 2026

Every Cure is a nonprofit that uses Google Cloud to match existing drugs against recognized diseases and accelerate drug repurposing research. The platform compares millions of drug-disease combinations using a unified biomedical knowledge base that brings together published research, clinical data, and molecular biology. Scientists and physicians review the top-ranked predictions to identify promising repurposing opportunities faster than traditional studies.

Organization: Every Cure
Industry: Healthcare
Location: United States
Published: July 2026

Reported outcomes

Experiments per year: 33.3×

Other quantified impact

Prediction pipeline duration: Approximately 17 hoursDrug repurposing programs launched: 9 count

Why do we believe this?Outcome claims, sources, and evidence checks

Normalized claim

Prediction pipeline duration: 17 hours decrease

Google Cloud Customer StoryJul 1, 2026Customer storyExplicit claimMedium evidence strength

compressed that cycle to approximately 17 hours

Normalized claim

Prediction pipeline duration: 99.3% decrease

Google Cloud Customer StoryJul 1, 2026Customer storyInferred claimMedium evidence strength

The earliest version of the platform took 100 days to complete a single run

Normalized claim

Experiments per year: 33.3 x increase

Google Cloud Customer StoryJul 1, 2026Customer storyInferred claimMedium evidence strength

allowed the team to manage only three experiments a year

Normalized claim

Drug repurposing programs launched: 9 count increase

Google Cloud Customer StoryJul 1, 2026Customer storyExplicit claimMedium evidence strength

launched nine active drug repurposing programs within one year

Why do we believe this deployment?Customer identity, provider attribution, maturity, and source checks

Customer: Every Cure
Provider: GCP
Maturity: Scaled Production
Linked source: Google Cloud Customer Story

Built a computational pharmacophenomics platform that evaluates all existing drugs against all recognized diseases at scale

Customer identity supportedSource describes one deploymentMaturity supported

Primary read

Use case focus

Showing 1 of 1

1Multimodal analytics

Built a computational pharmacophenomics platform that evaluates all existing drugs against all recognized diseases at scale.
Used BigQuery and Cloud Storage to unify biomedical knowledge graphs, then used Gemini Enterprise Agent Platform and Gemini to quantify evidence across the knowledge base.
Applied Google Cloud GPUs to shorten the prediction cycle and produce ranked candidates for expert review.

Technologies

BigQuery Cloud Storage Gemini Enterprise Agent Platform Gemini Google Cloud GPUs

Compressed a core prediction pipeline from about 100 days to about 17 hours.
Scaled experimentation from three runs per year to dozens of experiments per month.
Launched nine active drug repurposing programs within one year.
Enabled scientists, physicians, and PhDs to review the top 6,000 predictions over the past year.

Sources & evidence1

Evidence: Medium65/100Evidence strength