Fractional Data Architect on Thomas Nys | Fractional Data Architect

A plain statistical baseline caught model drift before any human would have.

Most pipelines are designed for the happy path.

A fintech founder asked me to sanity-check their data stack three weeks before a raise.

Three in four Belgian SMEs now use AI daily or weekly.

The lakehouse pitch is one platform for raw files and clean tables.

Your Databricks bill is mostly DBUs.

Praising the engineer who fixes everything at 2am rewards the exact dependency that eventually takes the platform down with them.

Streaming has a cost nobody puts in the business case: the weekly tax of running it after the demo works.

Investor technical due diligence has one real question: will this stack survive the growth the round is funding?

A wrong data engineering hire runs you about three salaries by the time you count the ramp, the re-hire, and the work that stalled in between.

You bought a semantic layer to end the “which number is right” argument.

Most data teams know their monthly cloud bill.

The 2026 data tooling is the best it’s ever been.

Run your AI feature twice on the same input.

Most teams collect everything “just in case,” then try to add privacy when an audit looms.

An e-commerce client’s reporting ran on a spreadsheet one person updated by hand every morning.

The best data engineers move for ownership and a manager who can describe the strategy.

The team bought more tokens.

The cheapest AI safety feature in 2026: “I don’t have current data.”

Nobody chose to keep 7 years of event logs.

The cloud migration finished in March.

The junior left in March.

Your pipeline failed at 3am.

Resource Monitors are uncomfortable on purpose.

The DORA metrics work for data teams.

Two teams sitting in different Slack channels build two pipelines that compute the same metric.

NIS2, DORA, and GDPR each want an incident report.

The six-hour batch was eating one engineer’s weekend every two weeks.

87% of tech leaders rate senior data engineer hiring as difficult.

If your post-mortems start with “who shipped that,” your org culture is the bug.

The AI Act compliance deadline is August 2026.

Most cross-team data fights are caused by an undocumented assumption upstream.

Two-engineer data teams keep picking Airflow because it’s “industry standard.”

A semantic layer is a consistency tool.

Software modernization and architectural modernization are different problems with different fixes.

Every data team has a Brent.

The day I realized our pipeline had become a person’s full-time job.

Your cloud data bill grew last year.

An empty “consumers” field is the signal to deprecate the asset.

Your data incident post-mortems end with “be more careful.”

Your data platform is at stage 2.

Most teams buy a semantic layer the way they buy a data warehouse.

Data mesh is an org design pattern.

Everyone says “people, process, technology.”

You have 47 data quality issues.

Your data platform has 6 vital signs.

Hybrid work changed how your data team collaborates.

Consolidating from 15 tools to 7 in 2026.

Most data catalogs go stale within a quarter.

Choosing between 3 data platforms?