How Do We Do It?
Data Analytics
Understanding data comes from exploring both the data and the underlying generation process. Our experience with statistical and graphical tools, coupled with our exploration process, allow us to create context for your modeling before we start.
Data Engineering
A key part of Machine Learning or AI project is the amount and quality of the data available to model. We work with clients to pull and integrate their data sources with external data to enhance predictive capability of your models. We have experience with all types of complex data such NLP text, IoT data, high frequency trading data, and time series.
High Performance Computing
Data size is increasing and model complexity is growing, so it is more important than ever to understand the underlying computing architectures and their capabilities. We have built custom software-hardware solution combinations for onsite and cloud-based platforms.
Applied Artificial Intelligence and Machine Learning
Our experience with techniques such as with neural net enabled tools such as BERT, fundamental packages TensorFlow & PyTorch, and models such as Random Forest and xgBoost allow us to separate and identify the practically possible from the hype.
Model Deployment
Timely model deployment and updating is a combination of tool knowledge, architecture, and practical skill. We deploy all three whether your problem cycle is measured in years or microseconds.
Clients
Public Sector
Social Security Administration
- Race & Ethnicity
- Anti-fraud
- Return to Work proof of concept
- BLS missing data imputation
Center for Medicaid Medicare Services
- Natural Language Processing (NLP) proof of concept
Export Import bank of the US
- Aircraft financing risk models and consulting
Private Sector
