A team without ML experts needs a custom model on its data. What fits?
Why B is the answer
Agent Platform AutoML
AutoML reduces the expertise needed.
In detail
- AutoML builds a custom model with little ML expertise.
- Check the results against simple baselines.
Use case
An illustrative scenario, not a real incident.
A retailer with no ML team trains a demand model with AutoML on its own data.
See it in different ways
flowchart TD A["Agent Platform"] --> B["Model Garden: find models"] A --> C["Agent Search: search and RAG"] A --> D["Agent Platform AutoML: custom models"] A --> E["Agent Studio: prototype"]
mindmap
root((Offerings))
Everyday work
Gemini for Workspace
Gems
Enterprise
Gemini Enterprise
Agent Platform
Customers
Agent Search
Conversational Agents
Contact center
Agent Assist
Conversational Insights
Developers
Model Garden
Google AI Studioflowchart LR S0["Provide data"] --> S1["AutoML trains"] --> S2["Evaluate"] --> S3["Deploy"]
Key point: AutoML lowers the expertise needed.
Why the other options are wrong
- A. Hand-written Python on a single VM
- That needs ML skills.
- C. Custom chip design
- Chip design is unrelated.
- D. A longer system prompt
- A prompt does not train on your data.