Code RoomCoding agent retrieval caching
HardPrep Room Coding #3225

Coding agent retrieval caching

System designDistributed systemsSenior–Staff~50 min

Design the backend for an AI coding agent that answers questions and makes edits over a developer's large monorepo (millions of lines, thousands of files). The agent calls an LLM in a multi-step loop (search code → read files → propose edits → run tests), so a single user request can trigger many LLM calls and tool invocations, and latency/cost compound. Design the retrieval + agent-orchestration + caching system to keep this responsive and affordable at scale.

What a strong answer looks like

Clarify scale and constraints first. Propose a clean component breakdown, then go deep on the hard parts (data model, bottlenecks, consistency, failure modes) and name the trade-offs you are making.

Clarify5:30 left
Estimate5:30 planned
Design16:30 planned
Deep dive13:30 planned
Failure9:00 planned
0:00
Which questions mattered is sealed until you submit. Telling you now would just be handing over the edge cases.