It’s a privilege and a great responsibility to play a role in shaping the evolution of one of the most important technology waves we’ve seen in decades.

By Adam Sylvain

As COO of Surge AI, a large and rapidly growing AI post-training company that brought in more than $1 billion in revenue last year, Bradley Webb T’16 is focused on one thing: delivering high-quality data, at scale, so AI models can become infinitely more useful and powerful. 

Webb and his colleagues at Surge recognize this work as essential to achieving AGI, "artificial general intelligence,” or the point at which AI will mirror or surpass human cognitive ability. When, or even if, that new frontier is reached will depend on increasingly sophisticated AI models accessing extraordinary amounts of specialized training data. 

“With NVIDIA leading the way in compute, and companies like OpenAI, Anthropic, and DeepMind making terrific progress in AI architecture, data is the missing piece,” says Webb. “It’s what stands between where we are now and getting to a world where AI models are infinitely interesting, and funny, and able to help address humanity-scale problems like, ‘How do we solve cancer?” 

For as much promise as the technology has shown, Webb says there is a long way to go to reach that kind of AI future. In the near term, he predicts the biggest business investment in AI will be in model evaluation. While current AI models are effective at information recall and responding to defined tasks, unexpected prompts often lead to “hallucinations” that produce false or misleading information. 

“Amid this rush to deploy AI we haven’t seen the requisite investment in evaluation frameworks that will help businesses ensure their investment in AI is economically valuable,” says Webb. “In the next five years, enterprise is going to need to invest in answering the question, ‘To what extent and in what capacity can I trust this model to perform in the way I need it to?’” 

Google, Facebook, and Twitter alum Edwin Chen founded Surge in 2020 in an effort to answer that very question. Webb became Surge’s third employee when he joined in 2021. In five years, the company—which never accepted outside capital—has grown from relative anonymity to achieve a multi-billion-dollar valuation, outpacing its biggest rivals, including Scale AI, Turing, and Mercor. This year, Chen became the youngest member of the Forbes 400 list, with an estimated net worth of $18 billion. 

From the beginning, Webb was won over by Chen’s vision of combining the best software with high-level human expertise. The company’s commitment to providing high-quality, human-centric data has helped Surge stand out amid a growing marketplace of competitors. 

“By bringing in the most capable people to work on data problems that are uniquely suited to their capabilities, we can observe and control the quality of data that we deliver, which leads to better-performing models,” he says. 

Grappling with the power and constraints of data is nothing new for Webb. More than a decade ago, he was working in finance on the product team for an advanced Navy satellite program. Webb became fascinated by the intricate process required to load and offload massive amounts of data captured by orbiting satellites for high-fidelity GPS and a variety of other applications. The experience led him to pursue his MBA and a successful pivot from finance into tech. 

At Tuck, Webb built a strong foundation to explore his growing interest in technology. He landed a coveted summer internship with Facebook and spent much of his second year “learning by doing,” immersed in projects that allowed him to hone the practical skills he would need as a product leader. 

“I remember trying to build a model that would allow me to predict the outcome of the 2016 Democratic midterms based on Twitter data and a bunch of other signals,” recalls Webb. “I didn’t even know how to code when I started that project, but it became this vessel where I could learn how to use Python and prepare myself in a lot of ways for the work I wanted to do coming out of Tuck.” 

After graduation, he returned to Facebook as part of a team focused on building the AI infrastructure needed to respond to human feedback. This included work on the in-app payment service Facebook Pay (now called Meta Pay). It was a period of exponential growth for Facebook, which grew from less than 10,000 to more than 70,000 employees between 2015 and 2020. Well before ChatGPT and large language models (LLMs) became a dominant topic of conversation, Webb was witnessing firsthand the adoption of AI across Facebook’s entire suite of products. 

“It became obvious AI was becoming an unstemmable tide that would influence pretty much every aspect of business moving forward,” he says. 

Webb left Facebook in 2019 and spent two years as a product leader for AppFolio, an AI-powered software platform for real estate managers. Now entering year five at Surge, one of his most urgent priorities is building and maintaining a workforce that will allow the company to sustain its growth and meet rising demand from industry. 

“We’re still early in the early parts of the AI adoption curve,” says Webb. “It’s a privilege and a great responsibility to play a role in shaping the evolution of one of the most important technology waves we’ve seen in decades. It’s not hyperbolic to say that AI will impact our economy in a way that rivals what the assembly line did more than 100 years ago.” 

With these implications in mind, Tuck has rapidly increased its engagement in AI—both within and beyond the MBA curriculum. As just one example, Tuck’s Center for Digital Strategies continues to play a leading role in The Dartmouth AI Conference, now in its third year. Webb, a past speaker at the conference, sees tremendous opportunity for Tuck students and alumni preparing to lead in an AI-powered business world. 

“Tuck is an ideal environment for people to experiment. Whether you’re using AI to learn how to code, or to riff on copy for a marketing campaign, the barriers for building a prototype and testing out ideas have been lifted in a lot of ways,” he says. “It’s one thing to learn and talk about the theoretical applications of AI. My advice is to get out there and build something.” 

This story originally appeared in print in the Winter 2026 issue of Tuck Today magazine.