Google Cloud Software / SaaS France

Maki

Customer Service Automation

In a market where applications pour in faster than HR teams can review them, the challenge of recruitment has changed in scale.

Read the original story Explore all use cases

"What allows us to understand a candidate's potential is not their stated background, but the way they reason, adapt, learn, or react in situations close to those they will actually encounter in their professional environment," explains…

For a company founded in 2021, this trajectory testifies to the maturity and relevance of its approach, but also to its infrastructure's capacity to support international deployments. "Without the reliability and elasticity of Google Cloud…

Founded in 2021, the company built its solution on proven scientific methods, drawing notably from Harvard University studies and decades of research in work psychology. These works converge on a single conclusion: structured interviews…

Reported impact

Business outcome

Maki transforms recruitment by automating interviews using a combination of AI and Google Cloud services. Based on proven scientific methods, the platform conducts real-time…; In a market where applications pour in faster than HR teams can review them, the challenge of recruitment has changed in scale. To help companies hire the talent they truly need,…

Solution

In a market where applications pour in faster than HR teams can review them, the challenge of recruitment has changed in scale. To build a solution worthy of its ambitions, Maki structured its architecture around Cloud Run and Vertex AI. "We preferred the serverless approach over a more classic architecture based on Kubernetes because Cloud Run adjusts capacity much more finely, at… On the other hand, Vertex AI allows Maki to orchestrate all of its models, whether it be Gemini for automated vision or open-source models. Key components include Gemini models provides the reasoning or generation layer used for classification, orchestration, content generation, or analysis; Cloud SQL stores transactional state and application data for the operational workflow at low latency; Cloud Storage stores large files, lake data, or static assets that feed downstream analytics and applications; Cloud Logging observes application health, latency, and runtime behavior across the deployment; Cloud.

Solution architecture

8 components · 4 layers

Compute
- Gemini models Provides the reasoning or generation layer used for classification, orchestration, content generation, or analysis.
- Vertex AI Provides the reasoning or generation layer used for classification, orchestration, content generation, or analysis.
- Cloud Run Runs serverless or virtualized application services with elastic scale for production AI workloads.
Storage
- Cloud SQL Stores transactional state and application data for the operational workflow at low latency.
- Cloud Storage Stores large files, lake data, or static assets that feed downstream analytics and applications.
Serving
- Conversational Agents Provides the collaboration or conversational surface where users work with AI assistance.
Governance
- Cloud Logging Observes application health, latency, and runtime behavior across the deployment.
- Cloud Monitoring Observes application health, latency, and runtime behavior across the deployment.

Architecture clues

"We also needed an environment that allows for fast innovation and simplifies the integration of new models. And that is exactly what we found with Google Cloud: infrastructure that automatically adjusts to the load,…
To build a solution worthy of its ambitions, Maki structured its architecture around Cloud Run and Vertex AI. "We preferred the serverless approach over a more classic architecture based on Kubernetes because Cloud Run…
To guarantee data availability, security, and compliance (notably regarding GDPR), Maki also relies on Cloud SQL, while multimedia content generated during interviews (video extracts, visual elements, or audio…
We preferred the serverless approach over a more classic architecture based on Kubernetes because Cloud Run adjusts capacity much more finely, at the level of each request, which guarantees optimal resource adjustment…
relies on Google Cloud and a combination of AI models to automate recruitment interviews and assess candidate skills at scale. By combining scientific methods, real-time conversational agents, and fraud detection using…

Evidence from the source

"What allows us to understand a candidate's potential is not their stated background, but the way they reason, adapt, learn, or react in situations close to those they will actually encounter in their professional…

Founded in 2021, the company built its solution on proven scientific methods, drawing notably from Harvard University studies and decades of research in work psychology. These works converge on a single conclusion:…

In a market where applications pour in faster than HR teams can review them, the challenge of recruitment has changed in scale. To help companies hire the talent they truly need, Maki imagined a new approach: AI…

Maki transforms recruitment by automating interviews using a combination of AI and Google Cloud services. Based on proven scientific methods, the platform conducts real-time interviews, detects cheating, and provides…

To turn this vision into reality, Maki relied on a combination of AI models and Google Cloud managed services. The heart of the platform rests on a real-time conversational agent capable of conducting the interview,…

With this sophisticated AI architecture, Maki quickly realized that the quality of its assessments would not be enough to keep the platform's promises: it also needed an infrastructure as solid as it was scalable. This…

Transparency Evidence & methodology 70/100 reference value · 75% confidence · 1 source

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

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 value70/100

Problem clarity20/20
Measures whether the customer's operational problem is specific enough to compare.
Implementation depth20/25
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 impact0/25
0 metrics describe measurable impact.
Deployment scale10/10
Rewards evidence of a live or scaled deployment and meaningful integration scope.

Evidence confidence75%

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
6 direct source quotes retained.
Date provenance0/10
The publication date has no documented provenance.
Source transparency8/10
1 source reference document publisher, URL, authority, retrieval, and supported fields.
Implementation specificity6/10
Rewards source-backed components, data flow, quantified impact, and a specific outcome.

Traceability

Sources & provenance

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

Google Cloud
Maki customer story

First-party
https://cloud.google.com/customers/maki

Source type
Vendor-published customer story
Supports
- Customer
- Business problem
- Solution
- Products
- Solution components
- Data flow
- Integrations
- technical area
- use case category
- Business outcome
- outcome category
- Metrics
- Architecture
- evidence quotes

Date evidence

First archived Mar 14, 2026 · Internet Archive · Independent First Wayback Machine capture; confirms the page existed by this date, not when it was published.
Open date source
Sitemap last modified Mar 10, 2026 · Google Cloud · Publisher metadata Publisher-reported modification timestamp; not a publication date.
Open date source

Not separately sourced Industry, Region, Company size. Treat these normalized attributes as lower-confidence until a dedicated source is added.

Similar use cases

See all like this

Google Cloud Modified · Oct 9, 2026 Tally Group Software / SaaS Australia This rapid processing surfaces targeted insights in under 20 seconds, eliminating manual checklists. At utility billing platform Tally Group , call center agents previously navigated up to six applications to… Customer Service Automation Apigee API Management · BigQuery · Gemini platform

Google Cloud Modified · Sep 18, 2026 Telexpress Software / SaaS Taiwan Telexpress wanted to replace rigid IVR workflows with natural, low-latency AI voicebots and chatbots that enterprise customers could adapt to their own routing rules and service processes. Gemini 2.5 Flash Customer Service Automation Gemini 2.5 Flash · Gemini Live API · Google Kubernetes Engine

Google Cloud Archived · Apr 13, 2026 Trillet Software / SaaS Australia Trillet needed to automate complex enterprise calls in sensitive industries, but earlier model choices produced unreliable answers and its fragmented infrastructure made it hard to scale voice workflows predictably and… Achieves <1% error rate Customer Service Automation Cloud Load Balancing · Gemini · Google Kubernetes Engine