I lead a team of seven data and AI engineers at a 3,500-person global manufacturer. We build the platforms that let the business act on its data, and I develop the engineers on the team.

In nearly two years, my team and I have built a unified enterprise data lake on Databricks, consolidating several dozen disparate sources - structured databases, Confluence spaces, SharePoint directories, and internal wikis - into a single foundation the business can self-serve from. I’ve personally led or built ingestion for Marketing, Sales, Supply Chain, Finance, and Engineering. I led our BI consolidation onto Databricks, which freed our data scientist from patching deploy tools to instead launch the company’s first public-facing AI agent. I architected and shipped our first unified customer profile, with advanced segmentation. I also cut roughly $200,000 in annual run-rate cost across AWS ingestion, Databricks orchestration, and BI licensing through value-based source selection, retention policies, Spark job tuning, and BI vendor consolidation.

Background

I started in enterprise systems consulting at Deloitte, where the work was data conversion and test leadership on a large product-administration implementation. In my experience, that is the part of a program where the problems in the data show up. I spent eight years at TeamSnap, a youth-sports SaaS, in three roles: lead data and software engineer, senior product manager for marquee partner implementations, and finally data platform lead, building the dbt and BigQuery platform and the first directly monetized data product. After that I held two utilities-sector roles: analytics engineer at Colorado Springs Utilities, where transformer-loading analytics informed seven figures of avoided capital, and lead data engineer at E Source on NYSERDA’s Integrated Energy Data Resource, a statewide utility data platform, where I rebuilt lineage under regulatory scrutiny and coached the team to a 98% processing improvement on the project’s most expensive software line item.

In every role I have worked on both the platform and the decisions it feeds. I build the pipeline and then use it, and I model the data and then argue about what it says with the person who has to act on it (and, anecdotally, I can’t fully express how empowering it is to be able to handle all of the engineering required to build a data product and then immediately begin using it to solve a business problem or illuminate an opportunity). The field notes on this site are written from that point of view.

Leadership

What I’m best at is making space and opportunity for the people on my team to do their best work. On my current team, a Data Scientist has grown into our AI Engineering Lead in roughly a year, and a Data Engineer has become a Sr. Data Engineer who now leads the unified customer profile work. Earlier in my career, a PM I worked alongside is now a Chief Operating Officer; an entry-level Ruby engineer I hired is a tech lead at Shopify; a Director of Marketing I built closed-loop ad attribution for is now a VP of Growth. “Measuring” results is my instinct, and the best measure I’ve found is to keep tabs on those I’ve mentored and led.

I speak the language of the manufacturer I serve. Our team’s metrics live as a mieruka board, andon, and digital obeya - Lean operations practices applied to data operations. That lets me work directly with plant leadership, IT executives, and operators as well as analysts. I choose to stay hands-on, and on a team of seven part of the engineering work falls to me in any case. I still write the hard MERGE, read the Spark plan, and sit in the vendor negotiation.

Next role

I am looking for Director / Head of Data roles, or a VP of Data & Analytics seat at a company that wants its data leader personally engaged in the technical work, remote or in Colorado Springs. I do my best work in a few situations in particular. Perhaps the company is multi-business or acquisitive and needs its data estate consolidated and governed, or its platform bill has grown faster than the value it delivers. Or, perhaps the leadership team wants AI tied to outcomes and governed like any other production system.

Credentials

B.S., Management Information Systems, Carlson School of Management, University of Minnesota. Databricks Certified Data Engineer Professional. I write Python, SQL, and Ruby, and I work in Databricks, Delta Lake, Unity Catalog, dbt, BigQuery, and the AWS and GCP services around them.

Contact

kyle.ries@gmail.com · LinkedIn · GitHub

I also write about experiments I run on my own time, such as EV consumption profiling from interval-meter data, when they turn into something worth reading.