



Competency and Services
We developed and deployed machine learning models for estimating the probabilities of specific public health outcomes. These models are used to target and evaluate programs. In either case, a model run could generate 4 million predictions. Supporting a scheduled use case and an on-demand use case required the ability to use the same ETL logic and predictive models with two different triggers. Meanwhile, predictions produced through a daily batch process could be automatically ingested into other systems directly from the analytics database.

We built an AWS-native data mart, consolidating product, marketing, and financial data into a unified, governed model. As part of the engagement, we rationalized overlapping sources, eliminated redundant transformations, and added monitoring and alerting to ensure data accuracy and freshness. The result was a set of maintainable workflows that significantly reduced cycle time for answering core operational questions.
In partnership with one of the largest charter school networks in the country, we used AI to detect inconsistencies in hiring and retention practices. We were able to quantify the extent to which the network's talent management practices were aligning with its mission and goals. The work involved developing customized data infrastructure to support predictive models, and led to more effective and equitable talent management aligned with the network's mission and goals.

Have an AI engineering need?