LOCKEDIN / FORWARD-DEPLOYED ENGINEERING 101
Understand it.
Then build it.
You can be an experienced engineer and still be new to agents. Start with what each thing is and how it helps a person do useful work. Explore 48 original introductions across software, agents, cloud, security and delivery. Start here without a model key; each topic connects to deeper instruction and practice.
- 101UnderstandPlain language, visual walkthrough, vocabulary
- 102Build with guidanceOriginal lessons, examples and knowledge checks
- 103Practise and explainRunnable labs, failure cases and handover
- 01Choose the project boundary
- 02Connect inference
- 03Evaluate and operate
A cloud AI platform
What is Google Cloud AI and Vertex AI?
Google Cloud is a cloud platform. Vertex AI is its platform for building and operating machine-learning and generative-AI workloads. Gemini is a model family, not another name for Vertex AI. A model call is one part of a Google Cloud application alongside compute, storage, identity and observability.
FDEs need to place AI into the customer's existing Google Cloud estate and hand it over with an operating contract.
CONNECT IT TO WHAT YOU KNOW
Start with something familiar.
Separate the workshop from the engine: Vertex AI provides platform capabilities, while a selected model supplies inference for your application.
A sample running in a console does not prove a deployed worker has the right identity, region, quota or access to company data. Our exercise is an architecture rehearsal; the linked provider quickstart performs actual cloud setup.
Words worth knowing
- Project
- A Google Cloud grouping and control boundary for resources.
- Vertex AI
- Google Cloud's AI platform.
- Gemini
- Google's model family, offered through supported products and APIs.
ONE EXAMPLE / FICTIONAL CASE
Watch the work take shape.
A worker reads approved procedure excerpts and sends a bounded request to a permitted Vertex AI model endpoint.
- 01Choose the project boundary
- 02Connect inference
- 03Evaluate and operate
Choose the project boundary
Identify the project, region, service identity and approved data location before enabling a workload.
A deployment decisionFROM RECOGNITION TO REASONING
Produce a Google Cloud architecture note for the maintenance assistant.
- 01
Map identity, compute, inference, source data and review storage.
- 02
Use the official quickstart to identify required setup, without creating resources for this exercise.
- 03
List the representative tasks and access failures you would test.
A service map, prerequisites and acceptance cases.
Compare your reasoning when you are ready
The model endpoint does not replace application storage or authorization. Keep the source-data boundary explicit and validate the selected model in the actual deployment environment.
Keep notes in your own working copy. Moving through these cards does not record completion.
TURN THE IDEA INTO USEFUL WORK
You can now go deeper.
A sample running in a console does not prove a deployed worker has the right identity, region, quota or access to company data. Our exercise is an architecture rehearsal; the linked provider quickstart performs actual cloud setup.
Academy lessons use your existing training access. Free guides stay open.
Learn with the original creators
Start with the conceptual overview. Then use the current Vertex AI documentation below for implementation.