# Context Engineering & Coding Agents with Cursor Source: https://www.youtube.com/watch?v=3KAI__5dUn0 Recap page: https://rapidrecap.app/video/3KAI__5dUn0 Generated: 2025-10-09T00:31:57.722+00:00 --- ## Quick Overview Cursor is evolving software development by integrating AI agents directly into the workflow, moving beyond simple auto-completion to sophisticated multi-agent systems that can handle complex tasks like full code reviews and long-horizon planning, aiming to make coding feel less like toil and more like creativity. **Key Points:** - Cursor's AI 'Tab' feature reduced the 'Shown rate' by 21% while increasing the 'Accept rate' by 38% compared to the old model, indicating higher quality suggestions. - The evolution of coding agents has progressed from simple word prediction to handling complex, multi-step tasks by incorporating context engineering. - Cursor introduced 'Agent Layout' to manage multiple coding agents concurrently, allowing them to run in isolation (using Worktrees) to avoid conflicts when modifying the same codebase. - Codebase Search agents use semantic search and embeddings, enabling them to quickly and accurately find correct code across files and directories, outperforming simple grep commands. - The future goal is to make building software feel less like toil and more like creativity by empowering agents to handle tedious tasks, such as writing documentation or triaging bugs. - The underlying principle for better agent performance is providing the model with high-quality, intentional context rather than relying on complex prompt tricks. ![Screenshot at 00:08: Lee Robinson introduces the talk 'Context Engineering & Coding Agents with Cursor' to a large live audience at OpenAI DevDay.](https://ss.rapidrecap.app/screens/3KAI__5dUn0/00-00-08.png) **Context:** This presentation, delivered by Lee Robinson (VP of Developer Experience at Cursor) and later by another speaker (likely Michael, CEO, based on context), details the advancements Cursor has made in integrating AI coding agents into the development workflow. The presentation traces the evolution from basic auto-completion ('Tab') to more advanced concepts like 'Agent' and 'Multi-Agents,' emphasizing context engineering as the key to unlocking greater automation and productivity for software engineers. ## Detailed Analysis The presentation outlines Cursor's advancements in AI-powered coding agents, starting with the evolution of the 'Tab' completion feature. The new RL model for 'Tab' resulted in a 21% reduction in the 'Shown rate' but a 38% increase in the 'Accept rate,' signifying higher quality suggestions. The speaker contrasts this historical progression of programming from punch cards to GUIs, noting that AI-assisted coding is now accelerating this evolution. Key innovations include improved context engineering, allowing agents to use broader context (like entire codebases via semantic search) rather than just local text. The 'Agent' feature allows for more complex, multi-step operations, demonstrated by the Composer agent which can create and manage to-do lists by analyzing requirements and code context. Cursor also introduced 'Worktrees' to allow multiple local agents to run in parallel environments, isolating their work to avoid conflicts when operating on the same codebase. Furthermore, agents are being trained to use tools like 'grep' internally and are becoming better at long-horizon tasks and integrating with development workflows (like planning and code reviews). The ultimate vision is to shift the focus of software engineering from tedious execution to creative problem-solving, enabled by agents that can handle complex tasks autonomously while still allowing human oversight and judgment. ### Evolution of AI Coding - Punch cards/terminals in the 60s/70s - GUIs in the 80s - AI-assisted coding now - AI agents are becoming more powerful and accessible. ### Tab Feature Improvement - New RL model achieved -21% Shown rate but +38% Accept rate, indicating better quality suggestions. ### Context Engineering - Moving beyond simple next-word prediction to giving models high-quality, intentional context, often by leveraging semantic search across the entire codebase. ### Agent Capabilities & Worktrees - Agents can perform complex tasks like multi-file edits and planning; Worktrees provide isolation for running multiple agents locally without conflicts. ### What's Next (Future Vision) - Agents will handle longer tasks, integrate into workflows (like PR reviews), and eventually move toward fully autonomous coding, focusing creativity on high-level problems rather than boilerplate. ![Screenshot at 00:05: Title slide for the presentation: 'Adopting Context Engineering with Cursor' at OpenAI DevDay.](https://ss.rapidrecap.app/screens/3KAI__5dUn0/00-00-05.png) ![Screenshot at 00:09: Lee Robinson, VP of Developer Experience at Cursor, begins his presentation on the evolution of software building.](https://ss.rapidrecap.app/screens/3KAI__5dUn0/00-00-09.png) ![Screenshot at 00:41: Slide showing historical coding interfaces \(terminal vs. GUI\) to set context for AI evolution.](https://ss.rapidrecap.app/screens/3KAI__5dUn0/00-00-41.png) ![Screenshot at 03:05: Slide comparing the 'Old model' vs. 'New RL model' for the 'Tab' feature, showing improved Accept rate and reduced Shown rate.](https://ss.rapidrecap.app/screens/3KAI__5dUn0/00-03-05.png) ![Screenshot at 01:35: Slide outlining the talk structure: Tab, Agent, Multi-Agents, What's Next.](https://ss.rapidrecap.app/screens/3KAI__5dUn0/00-01-35.png) ![Screenshot at 04:34: Introduction to 'Composer,' an agent capable of handling multi-file edits and plan management.](https://ss.rapidrecap.app/screens/3KAI__5dUn0/00-04-34.png) ![Screenshot at 06:00: Slide introducing 'Codebase Search,' highlighting the power of semantic search over simple grep commands for contextual retrieval.](https://ss.rapidrecap.app/screens/3KAI__5dUn0/00-06-00.png) ![Screenshot at 08:18: Slide titled 'Everywhere you work,' showing agent interaction across different interfaces \(CLI, Web\).](https://ss.rapidrecap.app/screens/3KAI__5dUn0/00-08-18.png) ![Screenshot at 10:48: Slide titled 'Commands, Rules, Hooks,' detailing extensibility features for agents.](https://ss.rapidrecap.app/screens/3KAI__5dUn0/00-10-48.png) ![Screenshot at 13:18: Slide titled 'Agent Layout,' introducing the UI for managing multiple agents concurrently for better workflow isolation \(Worktrees\). The toggle for 'Run local agents in separate worktrees' is visible.](https://ss.rapidrecap.app/screens/3KAI__5dUn0/00-13-18.png)