# CodeNib: A Multi-View Data System for Serving Repository Context to Coding Agents Page: https://stenobird.com/podcast/daily-paper-cast-7079649/codenib-a-multi-view-data-system-for-serving-repository-context-to-coding-agents Text version: https://stenobird.com/podcast/daily-paper-cast-7079649/codenib-a-multi-view-data-system-for-serving-repository-context-to-coding-agents.md Podcast: [Daily Paper Cast](https://stenobird.com/podcast/daily-paper-cast-7079649) Published: 2026-07-30T03:48:17+00:00 Episode link: https://share.transistor.fm/s/35258158 Audio file: https://media.transistor.fm/35258158/0c7dc107.mp3 Processing state: not_requested JSON: https://stenobird.com/v1/public/podcasts/daily-paper-cast-7079649/episodes/codenib-a-multi-view-data-system-for-serving-repository-context-to-coding-agents Duration seconds: 1140 ## Resource 🤗 Upvotes: 63 | cs.SE Authors: Zhongming Yu, Hengjia Yu, Boqin Yuan, Shuting Zhao, Yizhao Chen, Aryan Dokania, Mihir Jagtap, Jiayu Chang, Yitong Ma, Yash Jayswal, Wentao Ni, Hejia Zhang, Zhaoling Chen, Gangda Deng, Jishen Zhao Title: CodeNib: A Multi-View Data System for Serving Repository Context to Coding Agents Arxiv: http://arxiv.org/abs/2607.25431v1 Abstract: Coding agents repeatedly search, navigate, and retain context from evolving repositories, but disconnected indexes, language servers, and task-local histories force repeated discovery and obscure lifecycle costs. CodeNib builds reusable lexical, dense, and structural views per repository commit, maps outputs to repository-relative source ranges, maintains selected views across edits, and serves ranked search, symbol navigation, and bounded context through one runtime. Across 100 snapshots, we map quality-cost frontiers across the repository-context lifecycle. When outputs match an independent rebuild, graph and vector updates are $8.7\times$ and $25.4\times$ faster at the median. On the static-navigation subset matching normalized live-server locations (63% of 1,000 requests), the median per-request live/static latency ratio is $4.7\times$. Across five models, selected context policies preserve localization with 50--87% fewer trajectory tokens than paired grep/read. Together, these results support multi-view repository-context serving with explicit, operation-specific validity boundaries. ## Actions - request_transcript: `POST https://stenobird.com/v1/public/podcasts/daily-paper-cast-7079649/episodes/codenib-a-multi-view-data-system-for-serving-repository-context-to-coding-agents/transcription-requests` — Idempotently request low-priority transcript generation for this episode. - read_markdown: `GET https://stenobird.com/podcast/daily-paper-cast-7079649/codenib-a-multi-view-data-system-for-serving-repository-context-to-coding-agents.md` — Read the agent-friendly Markdown representation of this episode resource. A page view does not enqueue transcription. Agents should invoke `request_transcript` explicitly when they need this episode processed. ## Transcript Full transcripts are not published on public pages unless there is a clear rights basis.