Use Case Intelligence RSS Search use cases Search
Google Cloud Software / SaaS Australia

Atlassian

IT Operations Automation

Atlassian needed to upgrade its Rovo agent platform so it could handle enterprise-grade scale and quality, because lenient tool schemas in the legacy model were allowing subtle errors and limiting confidence in production rollouts.

Read the original story Explore all use cases
Boosted online agent task success rates to 80% with Gemini 3 Flash
Increased offline evaluation pass rates to 85% using Gemini 3 Flash
Scaled seamlessly to 200,000 daily invocations on Google Cloud
nearly 50% of customers to migrate within two weeks

Reported impact

Business outcome

Atlassian raised offline pass rates to 85%, boosted online agent success to 80%, moved nearly half of customers in two weeks, and scaled the platform to over 200,000 daily invocations in six weeks.

  • Customer Experience
  • Productivity Improvement
  • Scalability
  • Speed to Market

Solution

Atlassian migrated its Rovo personal assistant and agent platform to Gemini 3 Flash to improve both agent reliability and production scale. During evaluation, the team discovered that Gemini 3 Flash enforced stricter tool-schema behavior than its prior model, which forced Atlassian to tighten tool definitions and eliminate silent failures that had previously slipped through. The company then rolled the model out in phases, using an opt-in toggle so customers could switch agents gradually while offline and online evaluation results were tracked closely. This combination of schema refinement, controlled migration, and Google Cloud's resilient infrastructure let Atlassian scale from a single invocation to more than 200,000 daily invocations in six weeks while preserving the responsiveness needed for simple requests and deeper multi-loop reasoning for complex enterprise tasks.

Data flow

User requests enter the Rovo agent platform, are handled by Gemini 3 Flash against stricter tool schemas, pass through evaluation gates during phased rollout, and then return either quick responses or deeper multi-loop reasoning results depending on task complexity.

Solution architecture

2 components · 1 layer
  1. Compute
    • Gemini 3 Flash Powers the Rovo agent experience and enforces stricter tool schemas for higher-quality outputs.
    • Google Cloud Provides the resilient infrastructure that allowed Atlassian to scale agent invocations rapidly.

Architecture clues

  • Atlassian tracked both offline and online evaluations through the rollout.
  • Gemini 3 Flash forced Atlassian to tighten tool schemas that had previously been too lenient.
  • The migration was gated by an opt-in toggle rather than a big-bang switch.
  • The platform is designed to serve both high-speed simple queries and deeper multi-loop reasoning tasks.

Evidence from the source

Atlassian scaled seamlessly from a single invocation to over 200,000 daily invocations in six weeks
Gemini 3 Flash enforced stricter schema definitions, requiring Atlassian to update its tool definitions.
High satisfaction drove nearly 50% of customers to migrate within two weeks.
Increased offline evaluation pass rates to 85% using Gemini 3 Flash
This precision eliminated silent failures and significantly uplifted overall system quality.
Transparency Evidence & methodology 100/100 reference value · 89% confidence · 1 source

Review how this reference was scored and trace each normalized claim back to its documented source.

01

Scoring methodology

How scores are calculated

Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.

Customer reference value100/100
  • Problem clarity20/20
    Measures whether the customer's operational problem is specific enough to compare.
  • Implementation depth25/25
    Rewards named components, data flow, integrations, and architecture detail.
  • Outcome specificity20/20
    Measures whether the reported benefit is concrete rather than a generic claim.
  • Quantified impact25/25
    4 metrics describe measurable impact.
  • Deployment scale10/10
    Rewards evidence of a live or scaled deployment and meaningful integration scope.
Evidence confidence89%
  • Source authority20/25
    The main evidence is an official vendor-published customer story.
  • Claim traceability21/25
    10 of 12 key fields link to a named source.
  • Direct evidence20/20
    5 direct source quotes retained.
  • Date provenance10/10
    The publication date comes from the source page or publisher.
  • Source transparency8/10
    1 source reference document publisher, URL, authority, retrieval, and supported fields.
  • Implementation specificity10/10
    Rewards source-backed components, data flow, quantified impact, and a specific outcome.
02

Traceability

Sources & provenance

Field coverage shows which facts each source supports. Dates are shown only when documented by the source.

  1. Google Cloud

    Atlassian customer story

    First-party
    https://cloud.google.com/customers/atlassian
    Source type
    Vendor-published customer story
    Published
    Apr 22, 2026
    Supports
    • Customer
    • Business problem
    • Solution
    • Products
    • Solution components
    • Data flow
    • Integrations
    • technical area
    • use case category
    • Business outcome
    • outcome category
    • Metrics
    • Architecture
    • evidence quotes
    • Publication date

Date evidence

  • Published on source page Apr 22, 2026 · Google Cloud Press Corner · First-party
    Open date source
Not separately sourced Industry, Region, Company size. Treat these normalized attributes as lower-confidence until a dedicated source is added.

Related

Similar use cases

See all like this
Google Cloud Archived · Oct 16, 2026 Alta Software / SaaS United States Alta needed a transactable Google Cloud Marketplace path for its AI and technology spend management platform without pulling its product team off core forecasting work, while also helping finance and IT leaders… Saved 125 engineering and 60 project management hours IT Operations AutomationReal-time Analytics Cloud Pub/Sub · Cloud Run · Cloud SQL Google Cloud Modified · Oct 9, 2026 Anthropic Software / SaaS San Francisco, CA, USA During large-scale training runs, traffic is highly variable—dropping to near zero during model checkpointing before bursting to millions of events per minute. Anthropic needed a serverless messaging queue that scaled… Handles peak throughput of over 1M events per second Document ProcessingIT Operations Automation BigQuery · Bigtable · Cloud Pub/Sub Google Cloud Modified · Oct 9, 2026 Replenit Software / SaaS Poland Legacy marketing tools often treat individual shoppers as spreadsheet averages, relying on broad customer segments and predictable, rigid automation rules. As a result, brands frequently blast customers with irrelevant,… 360% business growth in one quarter with Google Cloud IT Operations AutomationMarketing Personalization BigQuery · Cloud TPUs · Gemini