How Google Cloud Empowers Developers: Agent Platform and RAG

Agent Platform is Google Cloud's unified environment for building with generative AI: one place where developers discover and evaluate models, adapt them with company data, build gen AI agents, and deploy and manage everything with enterprise-grade security. Instead of stitching together separate products for model access, search, training, and deployment, a development team works inside a single platform where those capabilities are designed to fit together.

The exam highlights three capabilities inside it. Model Garden is where developers find and compare foundation models. Agent Search provides ready-made, Google-quality search and question answering over an organization's own content. Agent Platform AutoML lets teams train custom models through a guided workflow rather than hand-built training code. Around these sit the tools to build custom agents — defining goals, instructions, and the data an agent can use — and to run them in production.

The business value is threefold. First, speed: teams start from pretrained models and prebuilt components instead of building from zero, so pilots become products in weeks rather than quarters. Second, accessibility: capabilities like AutoML and low-code agent building mean useful AI is no longer gated on scarce machine learning specialists. Third, governance: because building, grounding, and deployment happen on one platform, security controls, access policies, and monitoring apply consistently — which is what makes gen AI acceptable to enterprise risk and compliance teams.

Model Garden is a curated library inside Agent Platform where developers discover, evaluate, and start using foundation models. It brings together Google's own models — the Gemini family for language and multimodal tasks, Imagen for images, Veo for video, and the lightweight open Gemma models — alongside popular open models and models from Google's partners, each documented so a team can understand what a model does, what it costs, and what it is best at before committing.

The typical use case is model selection. A team building a customer-facing assistant can compare a frontier multimodal model against a smaller, faster, cheaper one, test both against real examples of their task, and choose based on quality, latency, and cost rather than guesswork. Because new and updated models appear in the same library, revisiting that choice later is straightforward.

The business value is choice without fragmentation. Model Garden prevents the two classic failure modes of enterprise AI adoption: betting everything on a single model that later proves wrong for the workload, and scattering different teams across disconnected tools with no shared governance. One library, many models, one set of enterprise controls — that framing is exactly what exam scenarios about Model Garden reward.

Agent Search gives an organization Google-quality search over its own data, out of the box. Developers point it at their content — internal documents, websites, knowledge bases, structured records — and it delivers a search and question-answering experience that understands natural language, ranks results by relevance, and can generate conversational answers grounded in the indexed content, with references back to the sources it used.

Use cases cluster in two areas. Inward-facing: an employee help center where staff ask questions in plain language and get answers drawn from HR policies, IT runbooks, and internal wikis instead of hunting through folders. Outward-facing: product search and self-service support on a company website, where customers get direct, sourced answers rather than a list of links.

The business value is time to value. Building a modern retrieval system in-house means solving document parsing, indexing, ranking, and answer generation — deep specialist work. Agent Search packages that entire stack as a managed offering, so the differentiating work left for the team is curating the content, not engineering the search. For a leader, the signal in an exam scenario is a company that wants trustworthy, sourced answers over its own information quickly and without a search engineering team — that points to Agent Search.

Agent Platform AutoML lets teams train custom machine learning models on their own data through a guided interface, without writing training code or designing model architectures. The developer supplies labeled examples — images, text, or structured records — and AutoML handles the model building, training, and evaluation automatically, producing a custom model tuned to that specific business problem.

It fits when a pretrained foundation model is not specialized enough and the organization has its own labeled data. Typical examples: a manufacturer training an image model to spot defects unique to its production line, a retailer classifying products against its own category tree, or a support team routing tickets using labels that only exist inside its workflow. In each case the knowledge that matters lives in the company's data, not in any general-purpose model.

The business value is democratized model building. AutoML converts a project that once required a machine learning research team into one a capable developer or analyst can deliver, which lowers cost, shortens timelines, and lets domain experts — the people who actually understand the data — drive the work. On the exam, the cue for AutoML is a scenario that needs a custom model from the company's own labeled data, delivered without specialist ML skills.

Retrieval-augmented generation (RAG) is the technique of retrieving relevant information from a knowledge source at question time and supplying it to a model so the answer is grounded in current, company-specific facts rather than only the model's training data. RAG matters because foundation models have a knowledge cutoff, know nothing about your private data, and can hallucinate — retrieval addresses all three without retraining the model.

Google Cloud offers two routes to RAG, and the exam expects you to tell them apart. Prebuilt RAG with Agent Search is the managed path: Agent Search handles the entire pipeline — ingesting documents, indexing them, retrieving the right passages, and grounding generated answers with citations — as one packaged capability. RAG APIs are the component path: individual services for the stages of a retrieval pipeline, such as parsing and chunking documents, generating embeddings, retrieving candidates, and ranking them, which developers assemble into a custom pipeline when they need control over each step.

The decision rule mirrors a classic build-versus-buy trade-off. Choose prebuilt RAG with Agent Search when the goal is grounded answers over company content with minimal engineering — fastest time to value, least to maintain. Choose RAG APIs when the team has specific requirements the packaged experience cannot meet — a custom retrieval strategy, unusual document formats, or tight integration into an existing application — and is prepared to own the pipeline it builds. Either way the business outcome is the same: answers users can trust because they are grounded in the organization's own sources.

Agent Platform is also where developers build custom gen AI agents — systems that use a foundation model to reason about a goal, then take steps to achieve it, rather than only answering a single prompt. Building an agent on the platform means defining what the agent is for, writing the instructions that shape its behavior and tone, connecting it to the data stores that ground its answers in company information, and giving it tools so it can act — looking up an order, filing a request, or calling another system.

Use cases go beyond question answering. A customer service agent can check a delivery status and initiate a return, not just explain the returns policy. An internal operations agent can gather figures from several systems and draft the weekly report. The common thread is multi-step work: the agent plans, uses its tools, and carries a task through to completion.

The business value of building agents on Agent Platform, rather than assembling the pieces independently, is that grounding, model access, deployment, monitoring, and security controls are already integrated. Developers focus on what the agent should do; the platform supplies how it runs, scales, and stays governed. Exam scenarios that describe a company wanting an assistant that takes actions or completes workflows — not merely answers questions — are pointing at custom agents on Agent Platform.

Use this table to match each offering to the problem it solves — most exam questions on this objective are exactly that matching exercise.

Notice the pattern: the offerings run from most packaged (Agent Search) to most flexible (RAG APIs and fully custom agents). Google Cloud's positioning is that a team should take the most packaged offering that meets the need — every step toward custom adds engineering cost that must be justified by a requirement the packaged option cannot satisfy.

Consider a mid-sized insurance company modernizing its customer and employee experience — a scenario shaped like the ones the exam presents. Its choices show how the offerings combine.

First, it deploys Agent Search over its policy documents, claims guidelines, and public help pages: customers get sourced answers about coverage on the website, and call center staff stop searching shared drives mid-call. This is prebuilt RAG — live in weeks because nothing had to be engineered. Second, its developers use Model Garden to select models: a frontier multimodal model for the complex claims assistant, and a smaller, cheaper model for high-volume email classification. Third, the claims team uses Agent Platform AutoML to train a custom image model on its own labeled photos of vehicle damage, producing a first-pass severity triage no general model could match. Finally, the company builds a custom claims agent on Agent Platform: grounded in the same policy data stores, able to check a claim's status, request missing documents, and draft the settlement letter — multi-step work, not just answers.

One platform carried all four projects, under one set of security and governance controls. When an exam question describes a business need, ask which layer it lives at — choosing a model, searching company data, training a custom model, or completing multi-step tasks — and the offering names itself.