A Career Between Academia and Industry Recognized
Mark Harman, a research scientist at Facebook and professor of Software Engineering at UCL Computer Science, has been elected as a Fellow of the U.K.’s Royal Academy of Engineering. The honor, which is by invitation only, recognizes his contributions across both academia and industry, particularly his work on search-based software engineering (SBSE), intelligent testing tools, and web-enabled simulation (WES). Harman is one of 50 engineers newly elected to the Fellowship this year.
Harman joined Facebook in 2017 after 25 years in academia, a transition he describes as a steep learning curve. “It’s been a baptism of fire, but I hope I am learning quickly,” he says. His academic work underpins much of his current industrial output. The SBSE approach, which applies search-based optimization algorithms to complex software engineering problems, forms the basis of Sapienz, a tool for automated code testing, and also feeds into WES, a project aimed at modeling and predicting harmful online behavior.
Before moving to Facebook, Harman directed the Centre for Research on Evolution Search and Testing from its founding in 2006. He maintains his part-time professorial role at UCL and hopes his new Fellowship will connect him with engineers beyond his own discipline, broadening his view of global engineering challenges.
From Theory to Deployment
Harman’s path to industry began with a growing frustration that his research was not seeing real-world uptake. “I was doing plenty of research but not seeing enough uptake,” he says. “I realized I was talking the talk on software engineering but perhaps not walking the walk.”
In September 2016, he cofounded MaJiCKe Limited with a team of engineers, building intelligent testing tools that identified defects in the Android operating system. That work evolved into Sapienz, which automated and accelerated testing processes that Harman calls “very tedious and error-prone.” When Facebook acquired Sapienz in 2017, Harman and two fellow MaJiCKe engineers, Yue Jia and Ke Mao, joined the company, seeing their research deployed at scale for the first time.
Simulating Platform Behavior
Harman’s current focus, WES, represents a deliberate shift in his research direction. “I realized that people can use software for good and for ill,” he says. “My career is now not just about making software better — it’s about intervening to make it easier for people to do good things and more difficult to do bad things.”
The project involves building a highly realistic, large-scale simulation of Facebook, using machine learning to train bots that mimic real user behavior. Two distinct bot communities have been created: one of “innocent” bots acting as potential victims, and another of “bad” bots that crawl the network seeking targets. Engineers use the simulation to understand complex interactions—including spamming and scamming—and to test countermeasures, such as limiting the volume of posts and messages bad actors can send.
The team’s near-term goal is to better understand how online communities form. In the longer run, they aim to apply SBSE techniques to automate the system’s processes. Harman hopes that within three to five years, advanced ML approaches will allow the team to anticipate harmful behavior before it occurs.
On Software and AI
Harman sees software as a uniquely malleable engineering material. “It’s the only form of engineering where the material you build with is completely virtual rather than physical,” he explains. This malleability, he argues, means software “does not have to be agnostic about the way that humans use it.”
He keeps a measured tone on the subject of artificial intelligence, avoiding sensational predictions. “There’s no magic involved,” he says. “So far, we’ve only seen software and algorithms finding patterns in vast amounts of data.”
Balancing his industry role with ongoing academic work, Harman values the combination of intellectual freedom and practical impact. “Academia is intellectually stimulating, but it can be unsatisfying when your ideas don’t come to fruition,” he notes. “In this situation, I am lucky to be able to have my cake and eat it.”
His career to date has earned him several distinctions, including the IEEE Computer Society’s Harlan D. Mills Award and ACM SIGSOFT’s Outstanding Research Award. Harman attributes his success to more than technical vision: “A vision or a north star is necessary but not sufficient. You also need a great team of people who share and understand the vision.”
