{"slug": "agentic-coding-in-the-wild-characterizing-github-copilot-traces-at-production", "title": "Agentic Coding in the Wild: Characterizing GitHub Copilot Traces at Production", "summary": "A new study from arXiv presents the first production-scale characterization of agentic coding workloads using GitHub Copilot traces from June 2026, covering 3.2 million users, 13 million sessions, 761 million LLM calls, and 95 trillion tokens. The analysis reveals that agentic coding sessions have sparse user turns with autonomous agent loops, KV cache hit rates averaging 90% within a turn but dropping to 55% across turn boundaries, and proposes a lightweight idle-time predictor capturing 86-90% of total idle time to improve resource orchestration.", "body_md": "# Computer Science > Artificial Intelligence\n\n[Submitted on 30 Jul 2026]\n\n# Title:Agentic Coding in the Wild: Characterizing GitHub Copilot Traces at Production Scale\n\n[View PDF](/pdf/2608.00101)\n\n[HTML (experimental)](https://arxiv.org/html/2608.00101v1)\n\nAbstract:AI coding agents like GitHub Copilot, Claude Code, and Codex interleave multi-step LLM inference with tool execution, creating a workload different from chatbots. We present the first production-scale characterization of this workload using sampled GitHub Copilot traces from June 2026, comprising 3.2M users, 13M sessions, 761M LLM calls, and 95T tokens.\n\nOur analysis reveals distinctive workload properties with important systems implications. For example, agentic coding sessions consist of sparse user-initiated turns, each unfolding into an autonomous agent loop of LLM calls almost always coupled with tool execution. This structure yields KV cache hit rates averaging 90% within a turn, but falling to 55\\% across turn boundaries and drastically invalidated after events like model switches or context compaction. Diverse workflows and user behaviors are observed with variable and long-tailed token consumption, time span, and tool calls. We highlight the difference between quick agentic turnaround times and the minutes-long user idle periods at turn boundaries, and design a lightweight idle-time predictor that captures 86-90\\% of total idle time, enabling proactive decisions for efficient resource orchestration. These findings challenge assumptions underlying current LLM-serving systems and provide an empirical foundation for agent-native infrastructure.\n\n### References & Citations\n\nLoading...\n\n# Bibliographic and Citation Tools\n\nBibliographic Explorer\n\n*(*[What is the Explorer?](https://info.arxiv.org/labs/showcase.html#arxiv-bibliographic-explorer))\nConnected Papers\n\n*(*[What is Connected Papers?](https://www.connectedpapers.com/about))\nLitmaps\n\n*(*[What is Litmaps?](https://www.litmaps.co/))\nscite Smart Citations\n\n*(*[What are Smart Citations?](https://www.scite.ai/))# Code, Data and Media Associated with this Article\n\nalphaXiv\n\n*(*[What is alphaXiv?](https://alphaxiv.org/))\nCatalyzeX Code Finder for Papers\n\n*(*[What is CatalyzeX?](https://www.catalyzex.com))\nDagsHub\n\n*(*[What is DagsHub?](https://dagshub.com/))\nGotit.pub\n\n*(*[What is GotitPub?](http://gotit.pub/faq))\nHugging Face\n\n*(*[What is Huggingface?](https://huggingface.co/huggingface))\nScienceCast\n\n*(*[What is ScienceCast?](https://sciencecast.org/welcome))# Demos\n\n# Recommenders and Search Tools\n\nInfluence Flower\n\n*(*[What are Influence Flowers?](https://influencemap.cmlab.dev/))\nCORE Recommender\n\n*(*[What is CORE?](https://core.ac.uk/services/recommender))# arXivLabs: experimental projects with community collaborators\n\narXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.\n\nBoth individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.\n\nHave an idea for a project that will add value for arXiv's community? [ Learn more about arXivLabs](https://info.arxiv.org/labs/index.html).", "url": "https://wpnews.pro/news/agentic-coding-in-the-wild-characterizing-github-copilot-traces-at-production", "canonical_source": "https://arxiv.org/abs/2608.00101", "published_at": "2026-08-04 19:09:40+00:00", "updated_at": "2026-08-04 19:22:33.665490+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "large-language-models", "ai-infrastructure", "ai-research"], "entities": ["GitHub Copilot", "Claude Code", "Codex", "arXiv"], "alternates": {"html": "https://wpnews.pro/news/agentic-coding-in-the-wild-characterizing-github-copilot-traces-at-production", "markdown": "https://wpnews.pro/news/agentic-coding-in-the-wild-characterizing-github-copilot-traces-at-production.md", "text": "https://wpnews.pro/news/agentic-coding-in-the-wild-characterizing-github-copilot-traces-at-production.txt", "jsonld": "https://wpnews.pro/news/agentic-coding-in-the-wild-characterizing-github-copilot-traces-at-production.jsonld"}}