Cursor is an applied research lab building extraordinarily productive human-AI systems. To start, we're building the engineer of the future: a human-AI programmer that's an order of magnitude more effective than any one engineer.

In this talk, we’ll explore what went into building the Cursor of today, and how we’re building for the future.

Recorded live in San Francisco at the AI Engineer World's Fair. See the full schedule of talks at https://www.ai.engineer/worldsfair/2024/schedule & join us at the AI Engineer World's Fair in 2025! Get your tickets today at https://ai.engineer/2025

About Michael
I'm building Cursor to speed up software engineering.

In the past, I've enjoyed working on statistical math research, LLM-driven recommendation systems, high-throughput drug pipelines, and popular programming competitions. I studied CS and Math at MIT.

Connect with me on github, linkedin, or at mntruell at gmail dot com!

About Sualeh
I’m building Cursor to discover a new way to write code. I owe much of my fun to my friends and MIT. I am extremely excited about the new capabilties of LLMs and applications to code tools.

Previously:

- Studied machine learning, number theory, performance engineering, and theater at MIT.
- Made early contributions to metaphor, an end-to-end LLM powered search engine (before it was cool).
- Represented Pakistan at the IMO 2016-2018. Studied and taught competitive math at the Pakistani math camps.
- Worked on translation at IBM Watson ML.

00:00 Introduction
00:33 What does programming look like in the age of AI?
01:38 Plugins vs Agents
02:59 Cursor’s first product: a code editor for programming with AI
03:30 Copilot++ : an autocomplete system that predicts your next move
04:11 CMD K: Instruction to code: select code and iterate with AI
04:35 Technical deep dive: Copilot++
06:19 New directions for the future (and tradeoffs)
09:12 Technical decisions when building Copilot++ & CMD K
12:48 Open positions at Cursor & who we’re looking for
14:10 Q&A: What are some more advanced techniques you’ve discovered beyond the obvious?
16:41 Q&A: Do you run personalized models? It feels like Cursor knows what I want to do next.

AI Engineer

The post discusses the development and goals of Cursor, an AI-integrated code editor designed to enhance programmer productivity. Founders Michael Truell and Sualeh Asif outline their vision for the future of programming in the AI age, focusing on tools that predict programmer actions and streamline code writing. They highlight features like the 'copilot Plus+' for predictive editing and 'command K' for transforming instructions into code. The ultimate aim is to create a tool that significantly automates software engineering tasks while keeping programmers in control.

Cursor: Building the Human-AI Hybrid Engineer: Michael Truell and Sualeh Asif