Loading…
Cursor
AI-powered code editor designed to make developers extraordinarily productive through intelligent coding assistance
- 126 posts summarised
- latest
The Good Stuff
Paid picks we use ourselves. Links carry a referral code — a subscription through one earns us a commission at no extra cost to you.
From their blog
Latest articles
The productivity impact of coding agents
The study examines how developers use Cursor’s agent and how its adoption affects software productivity across tens of thousands of users. After the agent became the default for eligible organizations, the rate of merged pull requests increased 39% relative to time trends in a baseline group, while revert rates did not change significantly, bugfix rates slightly decreased, and average lines edited and files touched per merged pull request were unchanged. Agent acceptance varied with experience: each standard-deviation increase in years of experience corresponded to an approximately 6% increase in acceptance relative to the mean, and experienced developers were more likely to plan before coding. In a sample of 1,000 users, about 61% of conversation-starting requests concerned implementation, with others focused on explanation or planning. The authors say no single metric yet captures AI’s economic impact on software engineering.
Improving agent with semantic search
Cursor describes semantic search as a complement to grep for helping coding agents locate relevant code in large codebases. Its evaluation used Cursor Context Bench across widely used models, comparing tool sets with and without semantic search, and found an average 12.5% accuracy improvement, with gains ranging from 6.5% to 23.5%. In an online A/B test, semantic search increased code retention by 0.3% overall and 2.6% on codebases with at least 1,000 files, while its absence produced a 2.2% increase in dissatisfied follow-up requests. The retrieval system uses a custom embedding model trained from agent session traces, with an LLM ranking helpful content and training aligning embedding similarity scores to those rankings. The conclusion is that combining semantic search with grep produces the best outcomes, especially for large codebases.
Introducing Cursor for Enterprise
Cursor is introducing enterprise features intended to make its coding agents more capable, transparent, and secure for organizations already using the product. Hooks expose the agent loop to custom scripts that can log actions and tool calls, enforce compliance or command allowlists, scrub sensitive material, connect external systems, and trigger automations; they can be distributed through MDM or Cursor’s cloud option. Team Rules provide shared context and can recommend or require standards such as API schemas, conventions, and workflows, while upgraded analytics report activity, adoption, AI lines of code, directory-group filters, and exportable data with updates every two minutes. The release also adds an Audit Log covering 19 event types and Sandbox Mode, which restricts network and file access by default, with administrators controlling availability and team-wide Git and network access.
Cloud Agents
Cursor’s Cloud Agents let developers run multiple agents remotely without keeping a laptop connected to the internet. Agents can be launched and managed from the Cursor editor, cursor.com/agents, Slack, Linear, and GitHub, supporting background bug fixes, small todos, and larger feature work. For difficult bugs, the team commonly has multiple models attempt the same problem and selects the best result; for complex features, a locally developed plan can be sent to a cloud agent for implementation, allowing work to continue on another task. Cursor also revamped its GPT-5 Codex agent harness for longer time horizons in the cloud, while plan mode supports sending implementation plans remotely. The workflow is available today in Cursor editor and Cursor Web, as well as the listed integrations.
Composer: Building a fast frontier model with RL
Composer is a mixture-of-experts agent model for software engineering that targets interactive use, with reported frontier coding results and generation speed four times faster than similar models on the authors’ benchmarks. It is trained with reinforcement learning on real-world software engineering challenges in large codebases, using production search and editing tools, terminal commands, and codebase-wide semantic search. Training responses may be code edits, plans, or informative answers, while the process rewards efficient tool use, parallelism, minimal unnecessary responses, and evidence-based claims. Evaluation uses Cursor Bench, built from real agent requests and curated optimal solutions, measuring correctness alongside adherence to existing abstractions and software engineering practices. The training stack combines asynchronous reinforcement learning with PyTorch and Ray, MXFP8 MoE kernels, expert parallelism, hybrid sharded data parallelism, and sandboxed cloud environments scaled to hundreds of thousands concurrently.
Introducing Cursor 2.0 and Composer
Cursor 2.0 introduces Composer, described as the product’s first coding model, alongside an interface designed for running multiple agents in parallel. Composer is built for low-latency agentic coding, completes most turns in under 30 seconds, and was trained with tools including codebase-wide semantic search to improve work in large codebases. The new interface centers on agents and outcomes rather than files, while still allowing users to open files in a redesigned layout or return to the classic IDE; parallel runs are isolated through git worktrees or remote machines. The release also targets review and testing bottlenecks by simplifying change inspection and adding a native browser tool that lets Cursor test its work and iterate, and Cursor 2.0 is now available.
Improving Cursor Tab with online RL
Cursor Tab predicts developers’ next actions across their codebases and handles more than 400 million requests per day, creating a large stream of accepted and rejected suggestions for training. The post describes replacing a separate acceptance filter with policy-gradient reinforcement learning, allowing the model to learn when to suggest and when to remain silent while reusing its learned code representation. An illustrative reward assigns 0.75 to an accepted suggestion, -0.25 to a rejected one, and 0 to showing nothing, targeting suggestions whose estimated acceptance chance is at least 25%; the production reward also considers suggestion size and jumps to other code locations. Because policy-gradient updates require on-policy data, each new checkpoint is rolled out to users; the current deployment-and-collection cycle takes 1.5 to 2 hours, and the resulting model makes 21% fewer suggestions with a 28% higher accept rate.
1.5x faster MoE training with custom MXFP8 kernels
An upgrade from Hopper H100s to Blackwell B200s exposed the Mixture-of-Experts (MoE) layer, implemented with MegaBlocks, as a major training bottleneck: it consumed nearly 53% of forward-pass time and 27% of backward-pass time. The team rewrote the layer at the GPU-kernel level using CUDA, PTX, and some ThunderKittens, while moving from BF16 to microscaled FP8 (MXFP8) with nearly zero loss in training quality. The implementation addresses Blackwell-specific tensor memory constraints and quantization overhead, including custom MXFP8 quantization that sustains more than 6.2 TB/s and emits scale matrices directly compatible with tcgen05.mma. On an internal model, MoE execution became 3.5x faster in both passes, producing a 1.5x end-to-end speedup on Blackwell and 2x versus the original Hopper setup, measured in tokens per second per GPU.
Bringing the Cursor Agent to Linear
Cursor has introduced a Linear integration that lets teams delegate issues to @Cursor by assignment or comment mention, then runs a background agent with issue details, comments, and linked references as context. Teams can also define triaging rules that automatically assign selected tasks and review agent diffs and pull requests from Linear. After connecting a Linear team through Cursor’s integrations page, users assign an issue or mention @Cursor; the agent analyzes it and notifies users when the work is complete and a pull request is opened. Cursor says its team uses the workflow for thousands of customer-feedback, feature, and bug tasks, including a product-bug and feedback triage rule. According to the post, the agent now drafts pull requests for thousands of issues and can one-shot many of them.
Updates to Teams and Auto
Two billing updates will apply at the next renewal after September 15 for monthly and yearly plans. Teams is moving agent usage from fixed request costs to variable request costs, so charges rise or fall with the amount of work performed; a simple syntax question costs less than implementing a full pull request. This change extends the credit system already used by individual plans and unifies billing across tiers. For individuals, Auto will count toward included monthly usage at competitive token rates at the next renewal after September 15, replacing its unlimited status since June 2025; the post says Auto had previously matched premium-model pricing for individuals from December 2023 through June 2025 and for teams since June 2025.
Cursor Agent CLI
Cursor has made its Agent available through a command-line interface (CLI) and in headless environments, extending it beyond the editor, web, mobile, and Slack. The release targets customers using Neovim, JetBrains, and other IDEs, while allowing Agent to run alongside those tools or as multiple parallel agents in a terminal or remotely. The CLI supports any model included with a Cursor subscription, and users can start it after installing with curl https://cursor.com/install -fsSL | bash and invoking agent chat with a prompt. Agent can generate files, run terminal commands, search a codebase, and now read, modify, or delete files and execute approved shell commands. The CLI is in beta, with evolving security safeguards and use limited to trusted environments.
GPT-5 is now available in Cursor
Cursor has made OpenAI’s GPT-5 available to users, presenting it as a coding-focused model based on an initial internal evaluation. Five engineers report using it for long-running tasks with Background Agents, parallel foreground agents, complex debugging, and Stripe-query latency optimization, while noting that explicit instructions improved its steerability. One engineer adjusted GPT-5’s verbose default style and follow-up behavior through rules. In a monorepo with separate frontend and backend submodules, GPT-5 built a backend API endpoint and corresponding React component, regenerated protobufs, and corrected resulting type errors in one shot. During launch week, paying users receive free credits, and the announcement characterizes GPT-5 as well suited to coding and tool calling.
Bugbot is out of beta
Cursor has released Bugbot, a code review agent that automatically analyzes pull-request changes for logic bugs, edge cases, and security issues before production. It runs automatically on pull requests, gathers context from changes to understand code intent, and comments detected issues directly where they occur. Teams can steer reviews with custom BUGBOT.md rules, then send flagged issues to Cursor or launch a background agent on the web with one click. During its beta, Bugbot reviewed over 1 million PRs and found over 1.5 million issues; more than 50% of identified bugs were resolved before merge. Cursor also reports adoption by several companies, with testimonials citing review-time savings, strong bug detection, and confidence reviewing AI-generated code.
Clarifying our pricing
Cursor acknowledges that its individual-plan pricing changes were communicated poorly and says the rollout damaged user trust. The revised Pro plan includes unlimited Tab and Auto usage, $20 of monthly frontier-model usage at API pricing, and the option to buy additional frontier usage at cost; a spend limit is available for users who do not want Auto. The company says it replaced a 500-request monthly allowance with usage credits because token costs vary substantially across models and tasks, while Auto routes requests among frontier models based on capacity. It clarifies that unlimited usage applies only to Auto, estimating the included Pro pool covers about 225 Sonnet 4, 550 Gemini, or 650 GPT 4.1 requests at median usage. Users with unexpected charges from June 16 through July 4 can request full refunds, alongside promised clearer notices, documentation, and dashboard visibility.
Cursor on web and mobile
Cursor announces that its Agents are available on the web and mobile, extending the IDE-based agent to desktop, tablet, and mobile browsers. Agents can run bug fixes, build features, answer complex codebase questions, and scaffold work in the background, while users can include images, add follow-up instructions, and run multiple agents in parallel. Repository-access team members can review diffs and pull requests or create pull requests from the web interface; the service can also be installed as a Progressive Web App on iOS or Android. Slack integration sends completion notifications and lets users trigger agents with “@Cursor,” and work can later be reviewed, extended, or edited inline in Cursor on a laptop.
Early team
Cursor introduces a representative sample of the people on its team and collaborators, spanning infrastructure, distributed systems, AI research, design, performance, and product development. The profiles include Aman Karmani, Jeremy Stribling, Jacob Jackson, Ian Huang, Sasha Rush, Alex Haugland, Ryo Lu, Lukas Möller, Jack Gallagher, and Michael Feldstein, with prior work at GitHub, Keybase, OpenAI, Cornell, Google, Notion, DeepJudge, Midjourney, Figma, and Meta. Their cited work includes KBFS, speculative decoding, a browser-based multiplayer Minecraft engine, language-model efficiency and long-context research, Cursor’s in-house inference engine, and Midjourney pretraining and reinforcement learning. The post invites people interested in transformational AI products, frontier RL, AI infrastructure, or distributed systems to contact [email protected].
A new Tab model
Cursor announces Fusion, a next-generation model for Cursor Tab, which predicts edits near the cursor and “jumps” to likely next editing locations. The post positions this as a response to the limits of cursor-only text insertion, with a broader goal of reducing code-editing tedium and moving toward in-flow Next Action Prediction. Compared with the March 2024 model, Fusion predicts over 25% more difficult edits per line, suggests over 10x longer change stretches, cuts median server latency from 475ms to 260ms, expands context from 5,500 to 13,000 tokens, and adds instant, accurate cursor jumps. These gains are attributed to cleaner and larger datasets, longer editor-state prompts, training for larger edits, synthetic instruction-following data, and inference and performance improvements. Fusion is rolling out to all users in client release 0.45.0.
Character Prefix Conditioning
The post poses a problem in code completion: generating a language-model completion that begins with the characters already typed by the user. Because modern models process token sequences rather than characters, naively tokenizing the input can produce the wrong result when the cursor falls inside a token. It introduces character prefix conditioning, defined as sampling tokens from an autoregressive distribution while constraining their decoded character string to start with the given prefix. The challenge is to construct an efficient autoregressive algorithm for this conditional distribution while minimizing calls to the original language model; the post requests either an algorithm description or an implementation.
Supermaven joins Cursor
Cursor announces that Supermaven, a fast, context-aware coding copilot, is joining the company, bringing creator Jacob Jackson and the Supermaven team into the effort. The post says the combination is intended to accelerate development of a more useful product built around purpose-built models, a tailored editing experience, and other product ideas. It anticipates product improvements, including a future version of the Tab model designed to be fast, context-aware, and highly intelligent during sequences of long changes. The Supermaven plugin will remain maintained, although Cursor will be the team’s core focus. The shift follows Supermaven’s earlier move toward an editor because extension APIs were blocking further capabilities the team wanted to build.
Iterating with shadow workspaces
Cursor’s shadow workspace addresses the problem of letting AI agents iterate on code without disrupting a developer’s active workspace. The design prioritizes LSP usability—especially lints, definition navigation, and language-server interaction—while also requiring independence, privacy, concurrency, universality, maintainability, and speed. A copied in-memory TextModel was rejected because sharing the user’s language server could introduce copied files into references, duplicate declarations, or miss errors. The implemented approach uses a reusable hidden Electron window for the workspace, routing AI edits through Cursor’s processes and returning lints independently of the visible editor. The post treats runnability as a harder future extension, with folder-proxy designs, platform limitations, and network-level independence still open questions.
Topics they write about
All topics →Weekly digest
The best engineering reads, once a week.