Agents And Tool Use easy
A teammate proposes splitting a single coding agent into five agents, arguing that more agents means smarter results. When is a multi-agent split actually justified?
- A Whenever a task is hard, since each added agent contributes extra reasoning capability.
- B Always, because agents that talk to each other catch every mistake the others make.
- C When subtasks need isolated contexts or can run in parallel without sharing state.
- D When the single agent hits rate limits, since each agent gets an independent quota.
Answer: C. When subtasks need isolated contexts or can run in parallel without sharing state.
Multi-agent splits buy **context isolation and parallelism**, not intelligence; each agent is the same model with a different prompt. The tradeoff is handoff overhead and lost context between agents. Splitting for hard tasks (A) often hurts, since a single agent with full context usually beats several agents passing partial summaries.
One system design question a day, with spaced repetition so the ones you miss come back.
Get the appRelated questions
- You are defining a 'schedule_meeting' tool for an agent. Which schema decision most reduces wrong or malformed calls?
- A model emits three parallel tool calls: `get_inventory`, `get_price`, and `get_shipping`. To cut latency, the host appends each result to the transcript as it arrives and immediately re-invokes the model. Runs start failing with malformed-request errors, and some produce hallucinated shipping costs. What should the host do?
- A team splits a coding agent into planner, coder, tester, and reviewer agents, and each receives the full transcript of the previous agent. The expected quality gains did not appear. Cost tripled, and the reviewer often fixates on approaches that earlier agents abandoned. When is a multi-agent split actually justified?
- A support agent should remember each customer's preferences across months of sessions. The team appends every past session summary to the system prompt. After six months, prompts exceed 80,000 tokens, latency has doubled, and the agent sometimes ships to an address the customer replaced weeks ago. What redesign fixes both problems?
- A data-analysis agent runs model-generated Python inside the same container as your API service. That container has database admin credentials in environment variables. A security review flags this as critical. What is the best mitigation?
- An internal agent has grown to 120 tools, many with overlapping names like search_docs, find_docs, and query_docs. Wrong-tool errors are rising. What is the best next step?