Code Less, Create More: Varun Mohan’s Vision for AI-Driven Engineering

n a riveting conversation on Lenny Rachitsky’s podcast, Varun Mohan, co-founder and CEO of Windsurf, unveiled a transformative vision for the future of coding, engineering, and learning.

Windsurf, an AI-powered Integrated Development Environment (IDE), has amassed over a million users in just four months, challenging competitors like Cursor.

Mohan’s journey—from GPU infrastructure to a groundbreaking coding platform—offers a glimpse into how AI is reshaping technology and the skills needed to excel.

Here, in third-person perspective, we explore Mohan’s insights, spotlighting key quotes that capture his bold predictions for learning, coding, engineering, and the critical role of agency.

From Infrastructure to Innovation: Windsurf’s Rise

Mohan’s entrepreneurial story began four years ago, when AI coding was unimaginable. He and his middle-school friend, with expertise in autonomous vehicles and Meta’s AR/VR, founded a GPU virtualization company to simplify deep learning for industries like finance and healthcare.

By mid-2022, their venture, Codeium, managed 10,000 GPUs, generated millions in revenue, and thrived with just eight employees. Yet, Mohan saw a bigger opportunity in AI-driven coding, inspired by tools like GitHub Copilot.

This led to Codeium’s pivot to an AI autocomplete tool, offered free across IDEs like VSCode, and later to enterprise solutions for giants like Dell and JPMorgan Chase.

Six months ago, Mohan made a daring move: forking VSCode to create Windsurf, a custom IDE built for AI’s potential.

“We realized we were getting limited by the IDEs we were working in,” Mohan explained, noting that Windsurf’s tailored interfaces tripled user acceptance rates for AI-suggested code refactors.

Coding’s AI Revolution: Engineers as Strategists

Mohan foresees a future where AI writes over 90% of software code, a prediction aligned with Anthropic’s Dario Amodei. This shift will redefine engineering, moving it from code-writing to strategic decision-making.

“AI is going to handle the vast majority, if not all, of the ‘solving it’ part,” Mohan said, referring to the act of coding once the problem and approach are clear.

Windsurf’s features, like custom review flows, enable developers to focus on overseeing AI outputs rather than writing lines of code.

AI will also tackle the “how” of coding by understanding company codebases and best practices. This leaves engineers to focus on the “what”: “What are the most important business problems that we need to solve?”

He envisions engineers as strategic thinkers, prioritizing product capabilities and making high-level technical choices, with Windsurf as their partner in this evolved workflow.

Learning Reimagined: Principles Over Code

As AI automates coding, Mohan believes learning must shift from syntax to problem-solving. Reflecting on his MIT education, he valued courses like distributed systems for teaching how to break down problems under computational constraints.

“Computer science is almost synonymous with problem-solving,” he noted, advocating for degrees that build mental models of systems—memory, processing, networks—over rote coding skills.

Yet, Mohan sees AI tools like Windsurf democratizing development. “It empowers a bunch of people that never understood all of those things to actually build as well,” he said, highlighting how product managers or self-taught coders can create software without formal training.

The future of learning, he argues, is about mastering adaptability and core principles, not memorizing frameworks.

Agency: The Skill That Defines the Future

For Mohan, the most critical skill in an AI-driven world is agency—the drive to identify problems and act independently. He critiques systems that reward following predefined paths, like completing assignments or executing narrow tasks.

“We don’t prioritize how to make sure you get people with real agency that want to build something,” he observed, noting its undervaluation in education and corporate hiring.

At Windsurf, agency is non-negotiable. “If we don’t innovate and do crazy things, we’re going to die,” Mohan declared, emphasizing the startup’s need to push boundaries.

AI amplifies agency, enabling builders to cross roles. A product manager at JPMorgan Chase, for example, could use Windsurf to edit code and push pull requests, becoming a force multiplier. Mohan sees agency as the trait that will set future builders apart, breaking down traditional role barriers.

A Call to Action: Dive In Now

Mohan’s advice to builders is urgent and practical: embrace AI tools like Windsurf and experiment relentlessly. “Get your hands as dirty as possible, as quickly as possible,” he urged, emphasizing the “tremendous amount of alpha” for those who master these tools first.

He encourages building apps, modifying codebases, and exploring new workflows, noting that many professionals remain unaware of AI’s potential.

Mohan’s philosophy embraces failure as a learning tool. “We just accept that we’re going to get a lot of things wrong,” he admitted, reflecting on Windsurf’s pivots and his wish to have acted faster. His team aims to cannibalize their product every six to twelve months, ensuring it evolves rapidly.

This mindset underscores his belief that staying ahead requires constant reinvention.

Shaping Tomorrow’s Tech Landscape

Varun Mohan’s journey with Windsurf—from infrastructure to a million-user IDE—mirrors the tech industry’s AI-driven transformation. Coding will become a strategic art, with AI writing most code. Learning will prioritize problem-solving over syntax.

And agency will define the builders who thrive. “There’s no ceiling at that point,” Mohan said of the opportunities for those who leverage AI to amplify their impact. For aspiring engineers, product managers, or founders, his message is clear: dive into tools like Windsurf, embrace the chaos, and shape the future.