Dataford
Interview QuestionsInterview GuidesExperiencesCoaching
New
Pricing
Get started
Dataford
Popular roles
Software EngineerData AnalystData ScientistData EngineerBusiness AnalystAI EngineerMachine Learning EngineerProduct Manager
Browse
Browse All RolesEvery role hub, from analyst to MLBrowse All CompaniesCompany-specific interview loopsAll Interview GuidesThe full guide library
Top questions by role
Software EngineerData AnalystData ScientistData EngineerBusiness AnalystAI EngineerMachine Learning EngineerProduct Manager
Top questions by skill
SQLPythonStatisticsMachine LearningA/B TestingSystem DesignGenerative AIProduct SenseMetricsBehavioral
Browse all questions →Try a mock interview
Experiences
Practice
Mock InterviewsTimed interview simulations with feedbackCoaching1:1 coaching with people at your target companiesSuccess PathYour 6-week structured planModulesCurated lessons by topicWebinarsTalks from ex-Big Tech data leadsPlaygroundA free-form scratch editor
Learn
BlogInterview strategy and career adviceTech Job Market ReportHiring trends across data and AI rolesFor UniversitiesDataford for career centersAbout DatafordWho we are and how we build
Pricing
Build my plan
Thread-Safe LLM Inference Cache
00:00
5 left

Thread-Safe LLM Inference Cache

MediumPython

Problem

Implement a thread-safe least recently used cache for frequently repeated Intone Networks LLM inference results. The cache stores string keys and arbitrary result values, returns cached results, and evicts the least recently accessed entry when capacity is exceeded.

Your function receives a capacity and a sequence of operations. Each operation is either ['set', key, value] or ['get', key]. Return the values produced by get operations in order. Return None for a cache miss. A set operation marks the key as most recently used, whether the key is new or already exists.

The implementation must protect shared state with a synchronization primitive so simultaneous get and set calls cannot corrupt the cache. Both successful lookup and update, including eviction, should run in average O(1) time.

Formal Specification

  • Input: capacity, a positive integer, and operations, a list of operations. Keys are strings, and values may be any object.
  • Output: A list containing one value for each get operation, in operation order.
  • Thread safety: Cache state must be protected while reading, updating recency, inserting, and evicting entries.

Constraints

  • 1 <= capacity <= 10^5
  • 1 <= len(operations) <= 2 * 10^5
  • Each operation is either ['get', key] or ['set', key, value]
  • Keys are strings
  • Values may be any object

Function Signature

def thread_safe_cache(capacity, operations):
Interviewer

Your question is Thread-Safe LLM Inference Cache. 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.

You need to log in / sign up to run or submit.
Code
Sign up free to run your codeLog inLn 2
Run your code to see test output here.