What challenges did UJJI AI face before AI implementation?

Traditional corporate training content production was slow and manual, relying on specialist workflows, approvals, and launch steps that took weeks. UJJI AI needed scalable, secure automation to create consistent multilingual training materials and summarize complex documents and videos for authorized users under strict regulatory requirements.

What was the impact of the AI solution?

Reduced time-to-launch for new learning paths from about two weeks to less than one hour. Accelerated training material development up to 300x. Cut production costs by 85%. Reduced repetitive tasks such as trial summaries and meeting minutes from 40 minutes to less than a minute. Reported total cost savings of about BRL 400,000 over three years.

Rank#74of 3,449

Source

GCPEvidence: Medium50/100

UJJI AI: Gemini Enterprise Agent Platform for automated corporate training material creation

Use case typeTraining infrastructure modernizationUpdated Jul 15, 2026

UJJI AI, based in Oxford and São Paulo, uses Google Cloud to automate the creation and updating of corporate training materials and internal documents. Its proprietary AI agent Liz uses Gemini Enterprise Agent Platform and RAG to analyze PDFs, training videos, and meeting recordings and generate structured learning paths.

Organization: UJJI AI
Industry: Education
Location: United Kingdom
Published: July 2026

Reported outcomes

−85%

production costsCost savings

300xtraining material development speed−99.6%time-to-launch for new learning paths−97.9%repetitive task duration400,000 BRLtotal cost savings

Strategic outcomes

Cost efficiencyAutomated creation of structured learning pathsRisk & complianceBuilt a governed platform for regulated financial customersOther strategic outcomeConverted a static repository into a live searchable knowledge library

Catalog median for cost savings deployments: −40% across 177 reported metrics. Compare benchmarks →

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

Normalized claim

Training material development speed: 300 x increase

Google Cloud Customer StoriesJul 15, 2026Customer storyExplicit claimMedium evidence strength

accelerated content development up to 300x

Normalized claim

Production costs: 85% decrease

Google Cloud Customer StoriesJul 15, 2026Customer storyExplicit claimMedium evidence strength

cut production costs by 85%

Normalized claim

Time-to-launch for new learning paths: 99.6% decrease

Google Cloud Customer StoriesJul 15, 2026Customer storyInferred claimMedium evidence strength

reduced time-to-launch for new learning paths from two weeks to less than one hour

Normalized claim

Repetitive task duration: 97.9% decrease

Google Cloud Customer StoriesJul 15, 2026Customer storyInferred claimMedium evidence strength

trial summaries and meeting minutes, which used to take 40 minutes, now take less than a minute

Normalized claim

Total cost savings: 400,000 BRL increase

Google Cloud Customer StoriesJul 15, 2026Customer storyExplicit claimMedium evidence strength

cost savings, totaling about BRL 400,000 over three years

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

Customer: UJJI AI
Provider: GCP
Maturity: Unknown
Linked source: Google Cloud Customer Stories

No explicit deployment-stage evidence found.

Customer identity supportedSource describes one deploymentMaturity evidence evaluated

Primary read

Use case focus

Showing 3 of 3

1Training infrastructure modernization
2Content generation
3Knowledge management

Traditional corporate training content production was slow and manual, relying on specialist workflows, approvals, and launch steps that took weeks.
UJJI AI needed scalable, secure automation to create consistent multilingual training materials and summarize complex documents and videos for authorized users under strict regulatory requirements.

Integrated its proprietary AI agent Liz with Gemini Enterprise Agent Platform using retrieval-augmented generation to analyze documents, videos, and recordings and generate structured learning paths.
Redesigned the platform on Cloud Run and Cloud Run functions for elastic pay-as-you-go scaling.
Added Cloud Armor and Identity Platform to protect sensitive information and enforce access control.
Used Cloud Storage to turn a static repository into a live searchable knowledge library.

Technologies

Gemini Enterprise Agent Platform Gemini Cloud Run Cloud Run functions Cloud Armor Identity Platform Cloud Storage

Reduced time-to-launch for new learning paths from about two weeks to less than one hour.
Accelerated training material development up to 300x.
Cut production costs by 85%.
Reduced repetitive tasks such as trial summaries and meeting minutes from 40 minutes to less than a minute.
Reported total cost savings of about BRL 400,000 over three years.

Architecture

A proprietary AI agent (Liz) is integrated with Gemini Enterprise Agent Platform using RAG to read PDFs, training videos, and meeting recordings. The platform runs on Cloud Run and Cloud Run functions for elastic scaling, uses Cloud Armor and Identity Platform for security and access control, and Cloud Storage for a large searchable knowledge library.

Sources & evidence1

Evidence: Medium50/100Evidence strength