Part 2 of 3 of the Meet Posh’s NLP team series. Even though I’ve lived in the Northeast for the past decade and a half, Michelle Fullwood — Principal Research Scientist at Posh — spots a tell-tale vowel pattern in my accent that tells her I grew up outside the region.Download the AI Checklist
, Boston-based Entrepreneur, Product Leader, and Startup Advisor
Even though I’ve lived in the Northeast for the past decade and a half, Michelle Fullwood — aPrincipal Research Scientist at Posh — spots a tell-tale vowel pattern in my accent that tells her I grew up outside the region. I quickly learn that Michelle’s deft identification shouldn’t come as a surprise; she is a linguist with a PhD in Linguistics from MIT, and a polyglot who has studied over a dozen languages to various degrees of fluency, including French, Mandarin, Japanese, Arabic, and Hieroglyphic Egyptian.
When you were young, what did you want to be?
It turns out that long before Michelle was working at cutting-edge startups and graduating from elite universities, she had quite a different aspiration: she wanted to be a cement truck driver. Yes. You read that right. This desire was so strong, that when Michelle’s younger sister was born, Michelle was given a cement truck toy to compensate for having to share her parents’ attention. It wouldn’t be until she was 14 years old that Michelle’s passion for language was kindled.
At age 14, Michelle found a copy of the Pelican Book classic Our Language by Simeon Potter, originally published in 1950. Her mother had won it as a school prize at the same age, but had found it dry and never finished it. She would never have guessed that this slim volume would change the direction of her daughter’s life.
Our Language dives into the delightful history of the English language, and it would become influential in Michelle’s educational decisions. Although she was on a scholarship to study mathematics at Cornell University, she decided to explore the field of linguistics as well, and fell in love with the subject. “It applied what I liked best about mathematics — finding and explaining patterns — to this deeply messy, uniquely human thing.” Little did Michelle know that her aptitude and passion were building her one of the most employable skill sets of the future. Michelle went on to work in NLP for the Singaporean Government and later returned to the US to pursue her PhD in Linguistics from MIT.
Before Posh, Michelle was working to better understand verbal abuse and harassment in online communities. Digital communities can be hunting grounds for scammers looking to prey upon unknowing victims. These bad actors most often prey on loneliness and financial insecurity, but defending community members at scale is notoriously difficult. Michelle identified tell-tale scammer communication patterns that could be detected long before they ever asked for money, then built a model to automatically flag these patterns, meaning the scammers could be stopped not only before they could pull off their scam, but even before the innocent community member formed an emotional bond with someone they thought was a friend. And as an added bonus, her technology doesn’t sleep. It’s always on watch. It’s basically an always-awake AI Batman.
At Posh, Michelle is thinking about scale. Many customer-bank relationships are similar across the country, and the conversations customers have with their banks or credit unions develop patterns. How can the NLP team at Posh take the conversations that millions of people are having with their financial institutions to improve the performance of a chatbot at even the smallest bank? Michelle spends a lot of energy and time discovering ways to generalize “customer intent” — what members mean when they ask questions. The value? Every institution Posh works with and every customer that engages with a Posh chatbot improves the system for everyone else.
“I like places that have a social good aspect,” she says before elaborating on the impact credit unions and community banks have on under-served communities that can’t get access to banking from large international banks. Empowering local banks and credit unions to deliver service that beats the biggest banks is a worthy cause.
Outside of social good, Michelle describes Posh as her ideal working environment. “The team is so smart, and people are very thoughtful. We aren’t just going after ‘cool’ things for the sake of cool.” While Michelle commends her team’s intelligence and thoughtfulness, she acknowledges some areas to improve; currently, the entire team is consciously and actively building a diverse and representative workforce.
I feel grateful having been able to meet and learn about Michelle’s journey and how Posh allows her to make an impact worthy of her talents. Her credentials are impressive, and it’s immediately obvious how smart she is. Fortunately for me, she’s hilarious, and she quickly grounded herself with funny stories of wanting to be a cement truck driver. In just one conversation, Michelle inspired me by her story of falling in love with language, pursuing the study of it, and now bringing about social good through its application.