Data & AI Engineer

Secretlab is an international gaming chair brand seating over three million users worldwide, with our key markets in the United States, Europe and Singapore, where we are headquartered.

This role is the natural next step of analytics engineering. The craft is the same: model the data properly. What changed is what consumes it. It used to be dashboards and analysts; now it's agents too, because people point agents at data to get analytical work done. So the models have to hold up under a machine reasoning over them, and you have to understand the mechanics of the agents doing the reasoning. Data engineering and AI engineering aren't two jobs stapled together here — each one sharpens the other. You work both from day one.

You ship real work from week one: real models, real loops, real users. Everything gets code review — everyone does, at every level. Your growth path is written down: the band above you has published expectations, and reviews are against those.

To be successful

• Ship correct data models. Staged layers, sensible grain, tested. Reviewers approve your PRs without re-deriving your logic.
• Build loops that take boring work off someone's plate — starting with your own. Every automation you ship can answer one question: how do you know it worked today
• Run AI-native. AI writes your first draft; you own what ships. It's a department expectation.
• Get better fast. Ask early, flag when something is bigger than it looks, and don't get the same review note twice.

What your week looks like

• Data engineering blocks — building and maintaining dbt models on Snowflake; code review both ways (you review too — it's how you learn).
• Loop building — automating a real task for yourself or the team on the platform, with a check that proves it ran right.
• Pipeline operations — ingestion issues, data quality flags, the unglamorous keeping-watch that keeps the data trustworthy.
• Weekly sync with the AI Engineering Lead on what you're building and what you're learning.

Requirements

The base — need it

• 2–5 years of relevant professional experience — data, analytics, or software work where someone depended on what you built. Strong side projects on top of that help; side projects instead of it don't.
• SQL you can defend live. Joins, window functions, why a slow query is slow. We test this.
• Data modeling fundamentals: why layers exist, what grain means, what breaks when you join carelessly.
• Enough Python to script real things, and working git habits.
• Already AI-native: you've automated real parts of your own work — study, projects, a job — and can explain why it's built the way it's built. Coursework alone doesn't show this; a system you actually use does.
• Can show how you know your work is right. Any method that produces evidence counts; confidence doesn't. It seemed fine is a no.

Bonus — moves you up the list

• dbt specifically (if your modeling is real, dbt comes fast).
• Snowflake, or any cloud warehouse.
• An orchestrator — Airflow or similar.
• A loop or agent with a real check on it — a test set, an assertion, anything that catches the bad day.
• AWS / Terraform / DevOps exposure. Genuinely a bonus: we do not expect it at this level and it won't block you.
• Anything shipped that someone other than you used.

Grow into — not expected at hire

• Cloud infrastructure and orchestration ops — the AWS/Terraform layer. Supervised at first; you'll pick it up on the job.
• Operating production agents — incidents, telemetry, the pager. That's the band above.
• Evals as a discipline — test sets, automated grading, regression gates. We hire for the instinct and teach the practice.
• Where this leads: both lanes stay open. Deepen the AI side and this role grows into a full AI Engineer. Deepen the data side and you're a full data engineer with a real AI base. Both paths are written down in the band above's expectations.

Personality

• Everyone here is held to the same five department expectations: run AI-native, own what ships, solve the business problem, say it straight, leave the process better than you found it. The role adds the technical bar on top.
• Wants both lanes. Some people want pure data engineering, some want pure AI. Both are fine roles. This role is both halves.
• Upfront and candid — honest about capability without embellishment; comfortable with direct feedback, in both directions.
• Pragmatic — asks what a request is for before building; doesn't gold-plate.
• Owns mistakes early — catches them before someone else does, and doesn't repeat them.

Explicitly not this role

• Prompt-tinkerers without an engineering base — can talk agents, can't write the SQL underneath them.
• Single-lane candidates who'd resent the other half of the week.
• Dashboard/BI-only profiles — this role builds the layer under the dashboards.