Alec Siems

Background

About Alec Siems

I am an award-winning data scientist who has developed predictive analytics models, dashboards, and web-based user interfaces to augment high value business processes and drive business AI growth. I am passionate about helping businesses increase productivity and save money by applying Artificial Intelligence (AI) and Machine Learning methods to raw data. I understand the most important factor for any AI/ML project are the business users and key decision makers. I consider myself a success when my customers feel that their work is more streamlined and their decisions are easier after embedding my AI/ML models into their processes. There are 3 key mantras I stick to when approaching any AI/ML project: 1. Simplicity is essential In the AI/ML space, it is very easy to overcomplicate a problem or work to build a solution that doesn't simplify the business process. I strive to provide relevant analytics, model interpretability, and visualizations alongside my models to give users proper context for decision making. This requires a broad skillset that includes business acumen, people skills (EQ), agile development, good communication, and the ability to work with a large range of technologies. 2."Every model is wrong, but some can be useful"- George Box Through the course of my career, I've had the opportunity to work with over a dozen large companies and equip them with strategies for achieving AI success. In many cases, I built web apps, dashboards, and models from scratch that allowed business users to interact with my work as much as possible. I quickly learned that users don't just need a forecast but instead need a way to interact with the model in a way that makes it more useful. Models don't live forever In fact, in my experience, they tend to 'die' rather quickly. The real world is complicated and there are always factors that are unaccounted for in the data. This is why the last part of an AI/ML project is the most essential and a primary focus for me in my projects: Model Operationalization. The data and model should be continuously monitored from several dimensions to ensure it is performing at acceptable levels over time. My focus is monitoring data drift, model accuracy, system performance, and usefulness. I work to automate as much of this process as possible so that I can continue to provide value through other projects. Please check out my experience and some of the blog post links included there. As always, feel free to reach out to me to discuss all things tech/AI/ML/business!