Caching LLM Responses | Dataford Interview Questions

Coding: implement a cache for LLM responses.

Asked in the Coding round stage. Listed as 'LLM cache' in the coding-round bullet.

Implement llm_cache(operations, capacity). Each operation is ['put', key, response] or ['get', key]. Return a list containing the response for every get, or None for a miss. A get marks the key as most recently used. When full, inserting a new key evicts the least recently used key. Updating an existing key also makes it most recently used.

Constraints

1 <= capacity <= 1,000
0 <= operations.length <= 1,000
Each operation is either ['get', key] or ['put', key, response]
Keys and responses are strings
Every key used by a get operation was previously inserted or may be absent from the cache

Function Signature

def llm_cache(operations, capacity):

Your question is Caching LLM Responses. Start with the requirements in the Question tab.

Run and submit as often as you like. When you're ready, talk me through your approach or go straight to the code.

Problem

Coding: implement a cache for LLM responses.

Asked in the Coding round stage. Listed as 'LLM cache' in the coding-round bullet.

Constraints

1 <= capacity <= 1,000
0 <= operations.length <= 1,000
Each operation is either ['get', key] or ['put', key, response]
Keys and responses are strings
Every key used by a get operation was previously inserted or may be absent from the cache

Function Signature

def llm_cache(operations, capacity):

Your question is Caching LLM Responses. Start with the requirements in the Question tab.

Run and submit as often as you like. When you're ready, talk me through your approach or go straight to the code.