Patrick Ma was building AI agents before the industry knew what to call them

The fear that AI will replace software engineers has become one of the loudest debates in technology. New tools can generate code, fix bugs, explain unknown files, and build simple applications with a few lines of natural language. To some people, it sounds like the beginning of the end for programmers. To Patrick Masenior AI engineer and Quora’s first dedicated AI agent engineer at Poe, sounds like a misunderstanding of what good engineers actually do. “AI can write code,” says Ma. “But it cannot take responsibility for the system.”

That distinction has shaped much of Ma’s recent work. At Quora, he built the first AI agent on Poe, the company’s AI product, before the software industry had settled on a common meaning for the word “agent.” He has worked on Poe for more than three years, transitioning from iOS engineering to agent AI work as the product and market evolved together. “I was building agents before ‘agent’ had a real definition,” says Ma. “That meant we couldn’t rely on buzzwords. We had to focus on what the product was going to do for users.”

Poe gives users access to leading AI models and lets them create AI-powered apps and experiences. Ma served as tech lead on major Poe launches covered by TechCrunch, including Previews, App Creator, and cross-model group chat. These products reflect a broader shift in software creation: more people can now describe what they want a computer to build, even if they don’t write traditional code themselves.

That’s one of the reasons Ma doesn’t believe the future is simply fewer engineers. He believes that the cost of building software is falling, which could increase the demand for useful software. The easier it becomes to start building, the more the industry needs engineers who can make systems reliable, maintainable and worth using. “Basic software tasks are much easier now,” says Ma. “But complex software still needs architecture. It still needs flavor. It still needs someone who understands what’s going to be built and how it’s going to fit together.”

This is where Ma’s point becomes more accurate than the common argument that AI is replacing jobs. AI can generate large amounts of code, but code volume is not the same as good engineering. A system must be reviewed, maintained, secured, evaluated and understood. It has to fit into a code base that other people can work with. It must behave correctly in critical situations.

“The best engineers won’t be the ones who avoid AI,” says Ma. “They will be the ones who use it aggressively without passing judgment.”

Mother speaks from unusual direct experience. He currently writes more than 99 percent of his code using AI. This does not mean that he has stopped as an engineer. This means that his role has shifted towards leading, reviewing, structuring and deciding.

“When I say I code with AI, I don’t mean I stop thinking,” he says. “I think more about the shape of the system, the constraints, and whether the output is actually correct.”

Inside Quora, Ma became a champion of AI coding tool adoption. In April 2025, he helped move from Cursor to Claude Code at a time when he says much of the industry had yet to recognize the wave of coding agents. His advocacy helped increase usage of the Claude Code by 20 times in two months.

That work wasn’t just about asking engineers to use a new tool. Ma also worked on making the technique itself more AI-native. He restructured parts of the codebase and even helped redesign the engineering interview process into an environment where AI tools became part of normal development. He also drove organizational adoption of agent evaluation tools.

“You can’t just give people AI tools and expect productivity to show up,” says Ma. “The code base, the process, the evaluation and the culture all have to change.”

It is one of the lessons he believes the industry is still learning from. AI adoption can become shallow when companies only measure usage. Ma has seen how token spending or leaderboard-style incentives can backfire. People can optimize for the metric instead of the work, even setting up pointless automated prompts to increase their AI usage numbers.

“AI usage is not the same as AI productivity,” he says. “If you measure the wrong thing, people will optimize for the wrong thing.”

The same caution applies to AI-generated code in serious systems. Ma believes that engineers should use artificial intelligence, but he also believes that they must remain responsible for what reaches production. He has seen cases where AI pushed bad code on behalf of engineers without sufficient human awareness. In critical systems, this may not become acceptable.

“AI agents are still early days,” says Ma. “They’re not reliable all the time. In real-world systems, people still have to review and take responsibility.”

That responsibility may become more important as software becomes easier to create. Poe’s coding agents are part of a larger movement toward letting users create software through natural language, but Ma believes accessibility doesn’t erase the need for engineering discipline.

A non-engineer might be able to generate a working app. An engineer still needs to understand if the code is secure, scalable, maintainable and aligned with the product’s true purpose. The difference between a fast output and durable software still matters.

“Software is not just about whether something runs once,” says Ma. “It’s about whether it continues to work, whether people can change it, and whether the system can survive real use.”

That belief also shapes how Ma evaluates the broader AI market. His work as a venture scout and hackathon judge has sharpened the way he separates useful AI projects from impressive ones. The question for him is not whether a demo will surprise. It’s whether the idea can become a system people trust.

“Whether you’re judging code or judging founders, you’re asking a similar question,” Ma says. “Is it real? Is it useful?”

His next role will bring this question even closer to the center of AI software development. Ma joins Cognition, creator of Devin, the first AI software engineer. At Cognition, his next chapter stays inside the same technical question: how much software engineering can AI take on, and what must still be governed by human judgment?

For Ma, the future of software engineering is a higher standard for using it well. Engineers may write fewer lines by hand, but they will need stronger judgment about architecture, product value, security and maintainability.

“AI lowers construction costs,” says Ma. “It also raises the bar for what engineers are responsible for.”