Job Interview - Data Analyst

Tell me about your background?

“I’m a chemical engineer by training who moved into operations and analytics because I enjoy turning messy, real-world data into clear decisions at speed. I started in capital project engineering at 3M managing nationwide projects, which taught me ownership, urgency, and execution. I now work as a Data Analyst at EnergyShrink, where I aggregate large operational datasets, build KPIs, and create dashboards used directly by leadership. I’m excited about Entyre Care because it operates at the same intersection of operations, speed, and meaningful real-world impact.”

Why Entyre Care?

“Your focus on speed, ownership, and immediate impact for families is exactly how I already work. Every role I’ve held has been execution-driven — delivering projects at 3M under aggressive timelines and building operational analytics at EnergyShrink. This role excites me because it’s not passive reporting; it’s about building systems, removing friction, and enabling fast, data-driven execution where the stakes are real.”

How strong is your SQL

My strongest hands-on work has been in Power Query, Python, and BI tools, where I’ve handled complex joins, aggregations, KPIs, and validation logic. I haven’t worked directly in Databricks SQL in production yet, but the logic I use daily maps directly to SQL. I’m actively deepening my SQL now and I’m confident I can become productive in Databricks quickly

Have you ever used Databricks?

Not in production yet. My experience so far has been across BI tools, Python, and structured operational datasets, but the workflows I’ve built — ingestion, transformation, validation, and reporting — are the same concepts used in Databricks. I’m very comfortable ramping quickly in new technical environments.

Describe a project you owned end-to-end.

At EnergyShrink, I took ownership of a project involving over 50,000 raw meter readings across hundreds of properties. I cleaned inconsistent inputs, standardized formats, built reproducible transformations, calculated KPIs, and delivered dashboards and leadership memos. The outcome was that leadership could immediately prioritize operational investments based on data instead of guesswork.

How do you ensure data quality?

“I focus on standardization, validation, and reproducibility. I use consistent formats, build logic that handles missing or inconsistent values, and verify outputs against source data. I also document each step so the process is auditable and repeatable. Reliable data is the only way operations can move fast with confidence.”

How do you work in high-pressure environments

“I’m most effective in high-pressure environments because I prioritize ruthlessly, focus on outcomes, and don’t wait for perfect information to act. My project engineering background trained me to move fast, manage ambiguity, and still deliver under deadline.”

What does ‘acting like an owner’ mean to you?

It means taking responsibility for the outcome, not just my task. If something breaks, slows down, or creates confusion, I treat it as my problem to solve — not something to escalate and wait on.

How do you balance speed vs accuracy?”

“I start with the minimum level of accuracy required to make a good decision quickly, then iterate. I don’t over-engineer early, but I also don’t release numbers I wouldn’t stand behind. Speed gets you momentum; accuracy earns trust.”

Why should we take a chance on you

Because I move fast, take ownership naturally, and already live in the intersection of analytics and operations. You won’t need to motivate me to act — only to point me at the most important problem.

Any questions for us?

• “What are the most critical metrics this role would own in the first 90 days?”

• “Where are your biggest operational bottlenecks right now?”

• “What would make someone in this role a clear success six months in?”

“What KPIs have you tracked?”

“At EnergyShrink, the KPIs I tracked focused on volume, efficiency, and quality. On the volume side, I tracked the number of properties and meters processed. On the efficiency side, I tracked energy intensity metrics like Energy Use Intensity and seasonal usage trends. And on the quality side, I tracked data completeness, validation pass rates, and consistency across sources.

What mattered most wasn’t just the metric itself, but making sure it was reliable, updated consistently, and directly tied to an operational decision. That same KPI mindset is exactly what I’d apply at Entyre Care — focusing on volume, speed, quality, and outcomes.”

How do you decide what should be a KPI?

“A KPI should directly tie to a decision. If no one would act differently based on that number, it’s not a KPI — it’s just a metric. I focus on what drives throughput, highlights bottlenecks, or signals risk.”

“What KPIs would you track in your first 30 days here?

“I’d start with throughput, backlog, and cycle time — how many families are being served, how long it takes, and where work is getting stuck. Those three immediately reveal where the biggest operational leverage is.”

“Walk me through your most complex data project.”

“At EnergyShrink, I aggregated over 50,000 raw meter readings across multiple energy types into standardized property-level datasets. I built normalization logic, validated seasonality, and created dashboards that helped identify high-impact optimization targets. The challenge was inconsistent formats and missing data, which I solved through structured cleaning rules, validation checks, and iterative QA with stakeholders.”

How would you build a daily SQL KPI report?

“I’d start from the raw operational tables, filter by date, apply business rules, aggregate to daily metrics, and create validation queries for row counts and anomalies. I’d structure it with CTEs for readability and automate refresh logic through Databricks or scheduled pipelines.”

What KPIs do you typically track?

“I track throughput, utilization rates, exception rates, time-based performance metrics, and trend deltas over time. At EnergyShrink these were property performance, usage variability, and anomaly rates; in BizOps, I’d focus on operational efficiency, service speed, and forecasting accuracy.”

Tell me about a time you had to move fast with imperfect data

At EnergyShrink, leadership needed a high-level scatter plot showing where schools currently stood relative to each other, but the dataset was incomplete and still had known quality issues. I moved forward by clearly defining what was directional versus fully validated, filtered out the highest-risk records, and annotated assumptions directly in the visualization. This allowed decision-makers to act quickly at a strategic level while I continued refining the underlying data in parallel. It balanced speed with transparency.”

How do you clean messy data?

“I identify a source of truth, standardize units and naming, flag partial records, and isolate anomalies rather than burying them. I treat data cleaning like engineering: systematic, testable, and repeatable.”

What is the difference between SQL and Python?

“SQL is for extracting, joining, and aggregating structured data efficiently. Python is for automation, forecasting, workflows, API connections, and advanced transformations that go beyond relational logic.”

Process improvement example?

“At EnergyShrink I streamlined data ingestion and reporting workflows so that datasets could refresh automatically with standardized outputs. This removed rework, reduced manual errors, and increased reporting turnaround speed.”