Normalized claim
Median setup time: 42 days
median setup times fell from 42 days to 3.1 hours
The Clinician no-code clinical project authoring wizard using Gemini Enterprise Agent Platform
Use case typeWorkflow automationUpdated Jul 12, 2026
The Clinician built TC-WIZ, a no-code project-authoring wizard powered by its Responsible AI Clinical Assistant (RAI) on Google Cloud. The system translates clinical protocol intent into validated platform configuration, including variables, cohorts, assessments, schedules, reporting logic, draft updates, scoring formulas, and recommended validated instruments. The architecture uses Gemini Enterprise Agent Platform (formerly Vertex AI), Gemini models, function calling, and Cloud Run functions with hybrid routing between Gemini Pro and Gemini Flash. The wizard includes PHI screening controls and licensing checks against curated instrument libraries.
- Organization
- The Clinician
- Industry
- Healthcare
- Location
- Australia
- Published
- July 2026
Reported outcomes
Median setup time: 42 days
Time & speed
Median setup time: 3.1 hoursMonthly project launches: 47 projects per monthMonthly project launches: 4 projects per monthUnauthorized instrument extraction attempts: 14 attempts
Planned next steps
- The source says the organization aims to achieve Annualized external consulting costs: 2.1 USD million.
Why do we believe this?Outcome claims, sources, and evidence checks
Normalized claim
Median setup time: 3.1 hours
median setup times fell from 42 days to 3.1 hours
Normalized claim
Monthly project launches: 47 projects per month
customers launched 47 new clinical projects through the wizard
Normalized claim
Monthly project launches: 4 projects per month
a massive leap from the legacy high of four projects a month
Normalized claim
Annualized external consulting costs: 2.1 USD million
customers avoid an estimated $2.1M in annualized external consulting costs
Normalized claim
Unauthorized instrument extraction attempts: 14 attempts
flagging 14 unauthorized instrument extraction attempts
Why do we believe this deployment?Customer identity, provider attribution, maturity, and source checks
- Customer
- The Clinician
- Provider
- GCP
- Maturity
- Production
- Linked source
- Google Cloud Customer Story
Deployed the system on Cloud Run functions with nine production endpoints and a hybrid routing pattern that sends complex reasoning and document extraction to Gemini Pro and high-volume tasks to Gemini Flash
Customer identity supportedSource describes one deploymentMaturity supported
Primary read
Use case focus
Showing 3 of 3
- 1Workflow automation
- 2Project management automation
- 3Healthcare workflow automation
- Built TC-WIZ, a no-code project-authoring wizard powered by RAI.
- Used Gemini Enterprise Agent Platform and Gemini models with function calling to update drafts, build form sections, generate scoring formulas, and recommend validated instruments.
- Deployed the system on Cloud Run functions with nine production endpoints and a hybrid routing pattern that sends complex reasoning and document extraction to Gemini Pro and high-volume tasks to Gemini Flash.
- Added PHI detection to block sensitive-data calls and embedded licensing controls to check requested clinical instruments against curated libraries.
- Median setup time fell from 42 days to 3.1 hours.
- Monthly project launches increased from 4 to 47 within 30 days.
- Customers avoided an estimated $2.1M in annualized external consulting costs.
- The system flagged 14 unauthorized instrument extraction attempts.
Architecture
TC-WIZ is a no-code project-authoring wizard powered by RAI and built on Gemini Enterprise Agent Platform (formerly Vertex AI). The implementation uses Gemini models with function calling, runs on Cloud Run functions, and routes work across nine production endpoints: Gemini Pro handles complex reasoning and document extraction, while Gemini Flash handles high-volume, latency-sensitive drafting tasks. The design also includes PHI screening and licensing controls tied to curated clinical instrument libraries.
Sources & evidence1
Evidence: Medium65/100Evidence strength
- Customer explicitly identified
- Deployment status explicitly supported
- Primary source available
- Quantified outcome available
- Technical implementation details available
Type: Customer StoryPublished: Jul 12, 2026Publisher: Google CloudEvidence: PrimaryConfidence: High
AI-generated summary. Verify important details with the linked sources before relying on this case.
Explore related AI use cases
Customer evidence
Provider evidence
Browse the catalog
Was this useful?
Community
Comments
No published comments yet.