Rob Brown, chief technology officer (CTO) at the Department of Homeland Security’s U.S. Citizenship and Immigration Services (USCIS), today discussed the importance of data quality checks as well as what steps USCIS is taking to accelerate its digital modernization.
During Cognilytica’s AI in Government event, Brown said he likes to provide human-in-the-loop aspects when it comes to the data quality that is being fed into machine learning AI models. This way, employees can enrich not only the actual data itself but also provide an eye for quality.
“In some areas, we actually have a team of humans, almost in parallel with some of the algorithms and automation in place and start to do cross-comparison to help enrich the data as well and help train or further train those models that are in play,” Brown said. “Those are some of the aspects that we’re doing from ensuring that we’re not only being fed good data but also doing quality checks on the data and enriching the data in that full pipeline.”
Additionally, Brown said it is important to have the right team of humans doing this work and stressed the importance of hiring and scaling up a skilled Federal workforce.
“We need to find good top talent. And by hook or by crook, trying to persuade them to join,” Brown said. “Ensuring that you continuously scale-up that workforce is also very, very important. We’ve seen great, huge leaps and bounds just doing that alone, but it needs to be focused and concerted. If it doesn’t have the right attention, then it will likely fail.”
As for other modernization efforts from the USCIS, Brown said, in his agency, he has seen “a definite trend towards automation, a definite trend towards the adoption, integration, and usage of AI and ML.”
“I still believe multi-cloud, is a ticket to ensure we’re getting the best out of the best. We’re in the business of immigration, we’re not in the business of creating past systems,” Brown added. “We’ve taken advantage of a lot of different cloud technologies, and will continue to do so, especially as it relates to moving the ball forward on AI and ML.”