A new Claude model release includes performance improvements for several reasoning tasks but has changed the format of its responses to system prompts that use multi-section instructions. Your application uses multi- section system prompts heavily. Initial evaluation on the application's actual workload shows the new model performs 8 percent better on reasoning tasks but produces malformed output on roughly 3 percent of requests because of the format change. The team is debating whether to upgrade.
How would you decide?

• A. Adapt the application's system prompt to the new model's format expectations and re-evaluate, then upgrade only if the adapted prompt eliminates the malformed output while preserving the reasoning improvements.
• B. Stay on the previous model permanently to avoid the malformed output rate and any future format changes that subsequent model releases might introduce in the application.
• C. Upgrade and add a downstream validation step that catches the 3 percent malformed output before it reaches users, treating the validation step as the team's mitigation for the format change.
• D. Upgrade immediately, because the 8 percent reasoning improvement outweighs the 3 percent malformed output rate across the application's typical request distribution.

Your Claude application has been running for several conversation turns, and you notice the model occasionally references information that was discussed many turns ago but is no longer relevant. You suspect context drift is causing the model to weight stale content too heavily.
How would you address the drift?

• A. Truncate the conversation so the model sees only the most recent turn during each subsequent response.
• B. Increase the context window size so all turns of the conversation remain visible to the model in full detail.
• C. Apply compaction to summarize older portions of the conversation so the gist remains while the specifics carry less weight.
• D. Reset the conversation after every turn so the model loses all prior turns when generating a response.

You are setting up a Claude application that will run a mix of multi-turn conversations and one-off requests.
You want to use caching techniques to reduce token costs where they apply. A teammate suggests caching the model's output as well, so the application does not have to make duplicate Claude calls when similar queries arrive.
You would apply prompt caching to...

• A. Only the user's input portion of each request because user input is the part of the prompt that varies the most across the application's normal operation.
• B. The static portions of prompts that are repeated across requests, such as system prompts, instructions, or shared context.
• C. Nothing, because prompt caching does not affect cost in any application that mixes multi-turn conversations and one-off requests in a single deployment.
• D. The model's output, treating the response from each request as cacheable content the application can return on similar future queries.

Your Claude application receives untrusted input from external sources. The team is establishing how the application should treat this untrusted input.
Untrusted input would be...

• A. Blocked entirely so the application accepts only trusted input, with no path for legitimate external input the application might process.
• B. Routed through a separate Claude application with more lax security controls so the original application does not handle the untrusted input directly.
• C. Treated identically to trusted input from internal sources, on the grounds that all input the application receives can be handled the same way.
• D. Validated and sanitized before being incorporated into prompts, and treated as data the model should not interpret as instructions.

You are designing a Claude application that maintains user sessions across multi-turn conversations. The product team has asked how the application will handle session lifecycle: when sessions should expire, how state is reset, and how the application avoids carrying stale context into new conversations.
How would you design session lifecycle?

• A. Define a single short session timeout that applies across all conversations and treat the timeout as the application's complete session lifecycle mechanism.
• B. Define explicit session expiration rules, state reset triggers, and rules for starting fresh sessions so stale context does not leak into new conversations.
• C. Define explicit session expiration rules but rely on users to start new conversations when they want fresh context, with no automatic reset triggers in the application.
• D. Define state reset triggers tied to specific application events but apply them across all sessions globally, with no per-session expiration rules.