SnowflakeExpert levelPage 3 of 3

SnowPro Specialty: Gen AI exam dumps: questions 201 to 287 of 287

Page 3 of the free SnowPro Specialty: Gen AI question bank for the SnowPro® Specialty: Gen AI exam. Questions 201 to 287 are listed below, the first 5 in full with answers and explanations. Back to page 1 for the exam overview and FAQ.

Question bank last updated April 2026

Start timed practice exam Questions 201 to 287

Free SnowPro Specialty: Gen AI practice questions

Questions 201 to 205 of 287

Pick an answer before you open the explanation. Each question also has its own page with a permalink.

SnowPro Specialty: Gen AI Question 201

Single answerDocker images

A data science team is deploying a custom inference service for a GenAI workload in Snowflake Container Services. The service uses a Docker image that includes Python dependencies, tokenization libraries, and model-serving code. During testing, deployments are slow and security review flags the image as unnecessarily large and difficult to patch. The team wants to improve startup time, reduce attack surface, and make future updates easier without changing application behavior. Which action is the BEST recommendation?

A
Rebuild the container using a smaller, production-focused base image and a multi-stage Docker build so only runtime artifacts are included in the final image
B
Keep the current image but add more CPU and memory to the container service so the larger image can be pulled and started faster
C
Bundle development tools, debugging packages, and test datasets into the same production image so operational troubleshooting is easier after deployment
D
Create a single monolithic image that contains multiple model versions and all optional Python libraries to avoid rebuilding images later

Show answer and explanation

Correct answer: A

Explanation

The best recommendation is to minimize the Docker image by using a smaller runtime base image and a multi-stage build. This aligns with standard container security and operational best practices: ship only the dependencies and artifacts needed at runtime, avoid build tools in production images, and reduce the number of packages that must be maintained and patched. In Snowflake Container Services scenarios, this is especially practical because image size affects deployment efficiency and maintainability. Candidates should recognize that Docker image optimization is not just about storage savings; it directly affects startup time, security posture, and operational simplicity for GenAI inference services. Relevant best practices are consistent with Docker guidance on multi-stage builds, minimizing base images, and reducing attack surface in production containers.

A. Correct.
Correct. A smaller base image combined with multi-stage builds is a standard Docker best practice for production workloads. Multi-stage builds let the team compile or assemble artifacts in one stage, then copy only the required runtime files into the final image. This reduces image size, lowers pull time, minimizes the number of packages that need patching, and decreases the attack surface. For Snowflake Container Services deployments, smaller images also help improve service startup efficiency because less data must be transferred and initialized.
B. Incorrect.
Incorrect. Increasing compute resources does not address the root problem: the image itself is oversized and includes unnecessary content. Image pull time and security exposure are primarily affected by image composition, not just by CPU or memory settings. While additional resources may help runtime performance in some cases, they are not the best remediation for slow deployments caused by bloated Docker images.
C. Incorrect.
Incorrect. Including development tools, debug packages, and test data in a production image is a common anti-pattern. It increases image size, expands the attack surface, and complicates vulnerability management. Troubleshooting needs should typically be addressed through separate debug images, logs, metrics, or controlled diagnostics rather than by shipping unnecessary tooling in the production runtime image.
D. Incorrect.
Incorrect. A monolithic image with multiple model versions and optional libraries makes the image larger, harder to patch, and more difficult to maintain. It also increases the risk that unused dependencies introduce vulnerabilities. Best practice is to keep images purpose-built and minimal, packaging only what is needed for the specific service version being deployed.

SnowPro Specialty: Gen AI Question 202

Single answerDocker images

A data science team is deploying a custom LLM inference service to Snowpark Container Services so it can be called from a Snowflake-based generative AI application. The service must start quickly, remain reproducible across environments, and avoid unnecessary security risk. The current Dockerfile uses FROM python:latest, installs build tools and Python packages in a single stage, and copies the entire project directory including notebooks and test data into the image. Which change would BEST improve the Docker image for this Snowflake deployment scenario?

A
Replace python:latest with a pinned base image version, use a multi-stage build to exclude build-time dependencies, and copy only the files required to run the inference service
B
Keep python:latest so the image automatically receives the newest Python patches, and add more package installation commands to ensure compatibility at runtime
C
Bundle model training code, notebooks, sample datasets, and debugging utilities into the same runtime image so the service can support both development and inference workloads
D
Use a larger general-purpose base image that includes common Linux utilities and compilers so troubleshooting can be performed directly inside the production container

Show answer and explanation

Correct answer: A

Explanation

For Docker images used in Snowpark Container Services, best practice is to build minimal, deterministic runtime images. In real deployments for generative AI inference, image size affects pull time and startup speed, and excess packages or files increase operational risk. Pinning base image versions instead of using latest supports reproducibility and controlled upgrades. Multi-stage builds are a well-established Docker best practice for separating build-time and runtime dependencies. Excluding notebooks, test artifacts, and training data from the runtime image aligns with the principle of least privilege and reduces the attack surface. These practices are consistent with Docker image hardening guidance and with Snowflake container deployment patterns that favor lean, production-focused images.

A. Correct.
Correct. Pinning the base image version improves reproducibility and avoids unexpected behavior caused by upstream image changes. A multi-stage build is a standard Docker best practice for removing build tools and intermediate artifacts from the final runtime image, reducing image size and attack surface. Copying only runtime-required files avoids bloating the image with notebooks, tests, and sample data, which helps startup time and operational efficiency in Snowpark Container Services.
B. Incorrect.
Incorrect. Using python:latest reduces reproducibility because the underlying image can change over time, which may introduce package conflicts or runtime regressions. While security patching matters, production images should usually be rebuilt from explicitly pinned, tested versions rather than relying on an unpinned latest tag. Adding more installation commands also tends to increase image size and complexity rather than solving the core issue.
C. Incorrect.
Incorrect. Combining training assets and development materials with the inference runtime image is a common anti-pattern. It increases the image size, slows image transfer and startup, and expands the security surface unnecessarily. In a production Snowpark Container Services deployment for inference, the image should be as minimal and purpose-built as possible.
D. Incorrect.
Incorrect. A larger base image with extra utilities and compilers can make ad hoc debugging easier, but it is not the best choice for a production inference service. It increases image size and attack surface and works against the goal of fast, reproducible, secure deployments. Production containers should generally include only what is needed to run the service.

SnowPro Specialty: Gen AI Question 203

Single answerSpecification files

A data engineering team is deploying a custom GenAI workload in Snowflake Container Services. They want the service definition to be version-controlled, portable across environments, and easy to review during change management. The team decides to use a specification file rather than defining the service inline in SQL. Which approach best meets these goals while aligning with how Snowflake uses specification files?

A
Store the service specification as a YAML file in a Snowflake stage, then reference that staged file when creating or updating the service.
B
Store the service specification only in a local developer workstation file system and rely on Snowflake to automatically sync it during service creation.
C
Convert the service specification into a Python worksheet and execute it so Snowflake can infer the service definition.
D
Embed the full service specification directly into every CREATE SERVICE statement because Snowflake does not support external specification files.

Show answer and explanation

Correct answer: A

Explanation

For Snowflake Container Services, specification files are used to define services in a structured, portable format, typically YAML. A best practice is to keep the specification file under source control and upload it to a Snowflake stage so it can be referenced consistently in deployment workflows. This improves auditability, promotes infrastructure-as-code style operations, and reduces errors from manually embedding large definitions inline. Options suggesting local-only files or Python inference reflect common misconceptions about how Snowflake consumes service definitions. Snowflake documentation for Snowpark Container Services describes using specification files and staged artifacts for service deployment.

A. Correct.
Correct. Snowflake Container Services supports service specification files, commonly written in YAML, and these can be stored in a Snowflake stage and referenced during service creation or update. This supports versioning, review, and portability across environments because the spec is managed as a file artifact rather than repeatedly hand-authored inline.
B. Incorrect.
Incorrect. A file on a developer's local machine is not automatically available to Snowflake. To use a specification file with Snowflake, it must be made accessible to Snowflake, typically by uploading it to a stage. Relying on local files breaks portability and deployment repeatability.
C. Incorrect.
Incorrect. A Python worksheet is not a substitute for a Snowflake Container Services specification file. Snowflake does not infer a service definition from arbitrary Python code. The service definition must follow the expected specification format rather than being reverse-engineered from worksheet logic.
D. Incorrect.
Incorrect. Inline service definitions may be possible in some workflows, but this option is wrong because Snowflake does support external specification files. Saying external specification files are unsupported is factually incorrect and contradicts the team's goal of using a reviewable, reusable artifact.

SnowPro Specialty: Gen AI Question 204

Single answerSpecification files

A data engineering team is preparing a Snowflake Cortex Analyst semantic model for a retail analytics chatbot. They want the model to answer questions consistently across business units, use business-friendly terminology, and avoid exposing raw table and column names such as SKU_ID or NET_SLS_AMT to end users. The team is defining a specification file for the semantic model before deploying it. Which action in the specification file best addresses this requirement?

A
Define logical business entities, metrics, and friendly field names in the specification so the model maps user questions to curated semantic concepts instead of raw physical schema names.
B
Store all warehouse sizing and auto-suspend settings in the specification file so Cortex Analyst can optimize how it interprets business terms.
C
List every possible natural-language question users might ask in the specification file so the model does not need semantic mappings.
D
Embed role-based access grants directly in the specification file so the model automatically hides unauthorized columns without relying on Snowflake security controls.

Show answer and explanation

Correct answer: A

Explanation

The best answer is to define a curated semantic layer in the specification file using business-friendly concepts that map to the physical schema. For Cortex Analyst and related semantic-modeling patterns in Snowflake, specification files are intended to describe the meaning of data so natural-language requests can be interpreted in a governed, reusable way. This includes defining logical entities, dimensions, and metrics with names that make sense to business users, rather than forcing users to know source-system column names.

The distractors reflect common misconceptions. Operational settings such as warehouse configuration are managed outside the specification file. Enumerating all possible user questions is not a practical or durable substitute for semantic modeling. And security remains the responsibility of Snowflake's native access-control framework; semantic specifications complement governance but do not replace authorization controls.

This aligns with Snowflake best practices for semantic modeling in Gen AI analytics use cases: separate physical schema design, compute configuration, and security administration from the semantic definitions that help AI systems interpret business intent accurately.

A. Correct.
Correct. The purpose of the specification file for a semantic model is to describe the business-facing semantic layer: curated entities, dimensions, measures/metrics, and user-friendly terminology that map to underlying data structures. This helps Cortex Analyst interpret natural-language questions using business concepts rather than exposing raw physical object names. In practice, this is how teams improve consistency, usability, and governance for conversational analytics.
B. Incorrect.
Incorrect. Warehouse sizing, compute management, and auto-suspend behavior are Snowflake operational settings, not semantic model specification concerns. They affect performance and cost, but they do not define business terminology or how natural-language questions map to tables and columns.
C. Incorrect.
Incorrect. Providing example questions can be useful in some AI workflows, but a semantic specification is not meant to replace structured semantic definitions with an exhaustive list of prompts. Relying on enumerating every possible question is not scalable and does not solve the core requirement of creating governed, reusable business mappings.
D. Incorrect.
Incorrect. Access control should be enforced through Snowflake security mechanisms such as roles, grants, and object-level protections, not embedded as a substitute inside a semantic specification file. The specification can describe semantics, but it does not replace Snowflake's native authorization model.

SnowPro Specialty: Gen AI Question 205

Single answerCreate compute pool

A data science team is deploying a containerized LLM inference service in Snowflake using Snowpark Container Services. The service must run on GPU-capable infrastructure, scale from zero when idle to control cost, and be created by an engineer who has been granted the minimum required privileges for infrastructure setup. Which action best meets these requirements when creating the compute pool?

A
Create a compute pool with GPU-capable instance families, set an appropriate MIN_NODES and MAX_NODES range that allows scale-down to zero, and ensure the engineer has the CREATE COMPUTE POOL privilege on the account.
B
Create a virtual warehouse with a larger warehouse size, enable auto-suspend, and grant the engineer CREATE WAREHOUSE because Snowpark Container Services uses warehouses for container execution.
C
Create a compute pool with CPU-only instances, configure the service to request GPU resources later, and grant the engineer OWNERSHIP on the target database because database ownership is sufficient to create compute pools.
D
Create a compute pool with GPU-capable instances and set MIN_NODES equal to MAX_NODES to guarantee elasticity, then grant the engineer CREATE COMPUTE POOL on the schema where the service will be deployed.

Show answer and explanation

Correct answer: A

Explanation

In Snowpark Container Services, containerized services run on compute pools. When a workload requires GPU acceleration, the compute pool must be created with a GPU-capable instance family. Compute pools are account-level objects, so the privilege required to create them is CREATE COMPUTE POOL on the account, not on a database or schema. For cost control, Snowflake best practices favor configuring pools to allow elasticity rather than pinning capacity unnecessarily. This question targets practical understanding of how infrastructure for Gen AI inference is provisioned in Snowflake: choosing the correct compute primitive, selecting appropriate hardware, and assigning the right privilege boundary.

A. Correct.
Correct. Snowpark Container Services services run on compute pools, not virtual warehouses. For GPU-based inference, the compute pool must use a GPU-capable instance family. To control cost, the pool should be configured with a node range that supports scaling behavior, including scaling down when no workload requires capacity. Creating a compute pool is an account-level operation, so the required privilege is CREATE COMPUTE POOL on the account.
B. Incorrect.
Incorrect. Virtual warehouses are used for SQL and many Snowflake compute workloads, but Snowpark Container Services runs containers on compute pools, not warehouses. Although auto-suspend is a valid warehouse cost-control feature, it does not apply to container execution in place of a compute pool.
C. Incorrect.
Incorrect. A service that needs GPUs cannot rely on a CPU-only compute pool and simply request GPU resources later; the underlying compute pool must provide compatible GPU-capable nodes. Also, database ownership does not grant authority to create compute pools, because compute pools are account-level objects.
D. Incorrect.
Incorrect. GPU-capable instances are appropriate, but setting MIN_NODES equal to MAX_NODES prevents elastic scaling across a range and does not support the stated goal of scaling down when idle. In addition, CREATE COMPUTE POOL is not granted at the schema level; it is an account-level privilege.

Timed practice exam

Take a SnowPro Specialty: Gen AI practice test under exam conditions

55 questions in 85 minutes, drawn from this bank, with a score report and a per-question review when you finish.

Start timed exam

SnowPro Specialty: Gen AI practice questions 201 to 287 of 287

Every question has a page with the answer and explanation. Numbers are stable, so you can bookmark or share them. The bank is split into 3 pages of up to 100 questions.