You land on https://www.cursor.so/ (Cursor) for the first time. What's unclear, what makes you leave, and what would make you sign up?
Page reviewed
AI Coding Agent for Building Ambitious Software | Cursor
https://www.cursor.so/ (opens in a new tab)
10 fictional AI personas. Exploratory perspectives, not a survey of real customers. Methodology & limitations
Simulated interest
AI synthesis, not measured demand
Suggested price range
No numeric price estimates in the answers. Ask a pricing follow-up to collect them.
Response sentiment
What this panel suggests
First-time visitors seem moderately interested in Cursor, but the landing page leaves enough ambiguity that many stay neutral rather than signing up immediately. The core promise is compelling: an AI coding agent that can turn ideas into code, work autonomously, and operate across tools. But respondents still want a clearer mental model of what that actually means in practice. In particular, they ask how much autonomy the agent has, whether it runs locally or in the cloud, and how much human oversight is expected. The page’s broad claims about agents working in the terminal, Slack, and GitHub help, but they do not fully answer workflow-fit questions for people using specific environments like IntelliJ or TypeScript/Python stacks. Pricing is another major blocker because the page does not make the commercial model obvious enough for visitors to judge whether it is worth trying. Security and specialized-environment fit also remain concerns. What would make them sign up is clearer proof: a concrete demo of autonomy, transparent pricing, and evidence that Cursor integrates cleanly with their existing setup.
Pricing & positioning
Do not treat the missing willingness-to-pay signal as rejection; instead, reduce uncertainty around signup by making pricing and plan structure explicit earlier in the journey.
No willingness-to-pay evidence was captured. The main pricing question is clarity: whether Cursor is per-user, usage-based, or team-based, and what is included at each tier.
Recurring themes
“What level of autonomy are we talking about?”Read Mert Mehlhorn’s response →
“I'm unclear on the pricing structure. Is this a per-user license, or based on usage?”Read Bessie Koss’s response →
“I'd leave if I couldn't quickly assess how it integrates with IntelliJ, my primary IDE.”Read David Olbrich’s response →
Objections to investigate
- Lack of pricing information
- Unclear technical implementation of agents
- Integration friction with existing stacks
- Security and data privacy concerns
Use these findings to plan customer interviews or experiments. How to interpret AI research
What the panel said
Click a name to open that persona’s full profile.
Bessie KossneutralThe core concept of an "AI Coding Agent" is intriguing, and the "Agents turn ideas into code" line immediately grabs my attention. I'm curious about how "Agents use their own computers" works – does this mean local or cloud resources? The demos showing interfaces in VS Code and Slack are helpful, but I'm unclear on the pricing structure. Is this a per-user license, or based on usage? Without more clarity on cost and how it integrates into a monorepo workflow, especially concerning security and data privacy in healthtech, I'd hesitate to sign up without more information. To sign up, I'd need to see transparent pricing and a clear explanation of the "agentic development" and "cloud agents" in relation to my existing stack and security requirements.
Simulated interest: 60/100
Jonathon HyattneutralLanding on Cursor, the core offering of an "AI coding agent" is clear enough, but the sheer volume of features and terminology like "agents," "Grok Bot," and "cloud agents" feels overwhelming. It's unclear how this integrates with my existing workflow or if it's just another layer of complexity. I'd leave because I'm wary of paid subscriptions and the "black box" nature of AI-generated code. My overhead is low, and I rely on free tools and documentation. To get me to sign up, you'd need to clearly demonstrate tangible cost savings or a significant time reduction that outweighs my current, carefully managed expenses, perhaps through a free tier that solves a specific pain point without demanding a subscription.
Simulated interest: 30/100
David OlbrichneutralFrom first glance, I appreciate Cursor's obvious focus on streamlining development workflows—something I wrestle with daily. The demo showing agents handling tasks like building dashboards is compelling, but I'm skeptical about the autonomous claims. My team's coding standards are stringent; I'd need proof that Cursor can adhere to our existing Kafka integration patterns and Java style guides without requiring constant oversight. The 'Download for macOS' call-to-action is clear, but I'd leave if I couldn't quickly assess how it integrates with IntelliJ, my primary IDE. What's unclear is whether it supports Kotlin beyond boilerplate or handles legacy system modifications—our biggest headache. The mention of 'handing off tasks' makes me nervous about overreliance on AI for production code. I might sign up for a demo if I could bring a specific incident response use-case (like our recent Kafka broker degradation) to test its triage capabilities.
Simulated interest: 40/100
Shantanu DhawanneutralThe core concept of an AI coding agent is intriguing, especially the idea of agents turning ideas into code and working autonomously. I'd leave if it felt too much like a black box or if the setup/integration process seemed complex and time-consuming, given our tight release cycles. What would make me sign up is seeing clear evidence of how it integrates with our existing TypeScript/Python stack and, crucially, how it impacts our software budget. If it genuinely accelerates development without introducing significant overhead or technical debt, and if there's a clear, affordable tier for a small team like ours, I’d definitely want to try it out. I'm curious about the pricing for teams and how it scales.
Simulated interest: 60/100
Armine PerezneutralThe core message, "Cursor is your coding agent for building ambitious software," is clear, and the concept of agents turning ideas into code is intriguing. What's unclear is the exact nature of these agents and how they integrate with my existing workflow in VS Code, especially regarding performance with large datasets, a key concern for my team. I might leave if the pricing isn't transparent or seems too high for a small squad without clear ROI. To sign up, I'd need to see concrete examples of how it handles complex Python/SQL data processing and React frontend tasks, and perhaps a free trial to test its capabilities against our specific challenges.
Simulated interest: 60/100
Erika OlsonneutralThe page states Cursor is an 'AI Coding Agent' which sounds powerful, but what exactly does that mean in practice? It mentions 'agents turn ideas into code' and 'works autonomously,' which is intriguing for someone juggling multiple projects. However, the concept of 'autonomous agents' running on their own computers is unclear. I'd leave if I couldn't quickly grasp the core functionality and see concrete examples relevant to solo developers. What makes me sign up is a clear demonstration of how it saves time for individuals like me, perhaps a trial that shows how it can accelerate tasks without requiring complex setup, and pricing that reflects value for a single user.
Simulated interest: 60/100
Mohammed BosneutralThe initial impression is that Cursor is an AI coding assistant, which is intriguing. However, the core functionality and how it integrates into a typical development workflow aren't immediately clear. Phrases like "agents turn ideas into code" and "works autonomously" sound promising for efficiency, but I'd need to understand the specifics of task delegation and review processes. What makes me leave is the lack of clarity on the practical application for existing projects and the potential overhead. To sign up, I’d need to see a clear demonstration of how it handles complex client portal or e-commerce integration tasks, and how it fits within a budget, perhaps with a trial that showcases these aspects specifically.
Simulated interest: 50/100
Duane MorarneutralLanding on Cursor's page, the main thing that's unclear is how the pricing works for the 'agents' and different models. It mentions various AI models, but without clear tiers or costs, it's hard to tell if it fits our nonprofit's budget. The concept of 'agents' building and testing features autonomously sounds powerful, but also a bit like a black box. What would make me sign up is a clear breakdown of pricing, especially a free or low-cost tier for small teams or open-source projects. Seeing specific examples of how it handles React or Node.js tasks we deal with daily, and maybe a trial period, would also be very convincing.
Simulated interest: 60/100
Greta ŁuczakneutralThe core idea of an AI coding agent that can "turn ideas into code" is compelling, especially the promise of handing off tasks. However, it's unclear how this integrates with our existing workflows. Specifically, how does it handle cross-platform consistency for iOS and Android, or our specific release scripts? The "Works autonomously, runs in parallel" feature sounds powerful, but I'd need to see concrete examples of its effectiveness in complex projects, not just high-level claims. For signup, I'd want to understand the pricing structure, especially for a small team, and see a clear path for how it would improve our current build and debugging processes without adding complexity. A trial or detailed case study showing impact on similar-sized studios would be persuasive.
Simulated interest: 40/100
Mert MehlhornneutralThe claims about "agents turning ideas into code" and "autonomous agents" are intriguing, but also quite abstract. What level of autonomy are we talking about? Can these agents handle the strict, safety-critical requirements of embedded systems, or is this more for typical web development? My work demands rigorous testing and validation, so I'd need to understand how Cursor integrates into that process. The idea of AI teammates is interesting, but I'd need to see concrete proof of reliability before trusting it with firmware that must function for years without failure. I'd consider signing up if there was a clear path to trial it on a non-critical embedded project, perhaps for test automation, and if its integration capabilities were transparent.
Simulated interest: 30/100
Keep exploring.
Find your next question or see which topics divide AI panels.
What would you ask next?
Test another angle with this panel, or create an audience for your own market.