Topics Practice Study

Zenaique

Personalized AI interview prep. Structured questions across every GenAI sub-discipline. Practice, score, master.

Practice

Topics
Practice
Challenges
Dashboard

Browse

Formats
Roles
Companies
Top questions

Product

Pricing
Sign in

© 2026 Zenaique

Home/Roles/Agentic AI Engineer

Agentic AI Engineer

339 questions

Builds autonomous, multi-step LLM systems: agent loops, tool calling, planning, memory, multi-agent orchestration, MCP, agent evaluation.

Sort

Questions

Fine-Tuning
Prompt Engineering
Context Engineering

Showing 1–25 of 339

← PrevNext →

What data shape does a tool calling fine-tune use, and what commonly goes wrong?

Short AnswerHard

Give one training example for SFT'ing a single weather tool call.

Short AnswerMedium

What does a JSONL row need beyond text SFT for OpenAI's hosted FT of tool calling models?

Short AnswerMedium

Describe the JSON shape of a Claude Opus 4.7 assistant message that calls a tool

Short AnswerMedium

Design a decay policy for an agent that has been talking to one user for two years

Short AnswerHard

What is the ReAct prompting pattern?

Where should an agent scratchpad live in the context, and what gets pruned from it each iteration?

FlashcardMedium

Why do plan then act agents keep an explicit plan block in context, even when the model could re-plan from scratch?

FlashcardMedium

How does LangGraph's typed state differ from a plain conversation buffer?

FlashcardMedium

Why are agentic systems shifting toward lazy retrieval via search tools?

FlashcardMedium

Why cap tool output size at the tool boundary instead of letting the model deal with it?

FlashcardMedium

How does streaming tool output complicate context budgeting in agent loops?

Compare episodic and semantic long term memory with examples from a personal assistant agent

FlashcardMedium

Why do hour scale agents use hierarchical summarization instead of a single summary block?

FlashcardMedium

When a parent agent hands off to a sub-agent, what is the minimum context it should pass and why?

Describe the three tier memory hierarchy used by modern LLM agent systems

Why does LangGraph checkpointing matter for long running context engineered agents?

FlashcardMedium

PremiumCompare eager memory injection…

FlashcardMedium

Order the four phases of a single iteration of a ReAct (Reason + Act) loop in the correct sequence.

Order StepsMedium

What does the model emit next in this ReAct trace, given the observation just returned?

Predict OutputHard

PremiumWhich is the right…

Multiple ChoiceMedium

PremiumWhich is the safest…

Multiple ChoiceHard

PremiumSpot the context engineering…

Spot the ErrorMedium

Select the parts of a long agent reasoning trace that are safe to prune from context between iterations

Multi-selectMedium

Match each memory framework to its core abstraction

Match PairsMedium

Voyage Ai Yellow Ai

Anthropic OpenAI

Braintrust Coinbase

Forethought Rephrase Ai

Character Ai Doordash

Modal Labs Salesforce

Baseten Inflection Ai

Braintrust Cohere

Accenture Haptik

Infosys Mu Sigma

Cursor LangChain

Inflection Ai Polyai