How does your data team unit test data transformation code, especially when that code is written in SQL?

Data & Analytics
2.6k viewscircle icon3 Comments
Sort by:
Head of Technical Delivery Services in Travel and Hospitality2 years ago

Use simulated data streams to create artificial datasets using data tranformation code and validate it against expected output.

Director of Customer Engineering - APAC in Software2 years ago

Validate with dummy data

Director of Engineering in Healthcare and Biotech2 years ago

Use fake input into the code under test and check to ensure the output matches what's expected. 

Content you might like

Among your organization’s leadership, how would you characterize expectations for GenAI?

Unrealistic13%

Highly optimistic48%

Moderately optimistic20%

Cautiously optimistic14%

Pessimistic4%

Other1%

View Results

How do you think AI will disrupt business across industries? Add to my list: 1. Content creation 2. Photos and video production 3. Basic coding and debugging 4. Strategic analysis to be highly complimented 

Read More Comments

Which pitfalls—model bias, false positives/negatives, data quality, regulatory constraints—often impede AI-based security tools, and how can they be mitigated in a financial-services context?

Read More Comments

AI and Research in Marketing and Communications: How Are You Ensuring Quality and Truth Over Speed? As AI becomes embedded in our day-to-day marketing operations, from customer insight to content strategy, it’s clear we’re gaining speed and scale. But are we sacrificing veracity and quality of output in the process? One of the biggest challenges I'm seeing is the rise of hallucinated content in AI-generated research.   - How do we ensure AI systems are surfacing trustworthy, contextually accurate, and well-sourced information? - Are we building enough verification layers into our AI-driven content and research workflows? - And how are you educating teams (and execs) to challenge AI outputs, rather than accept them at face value? - What frameworks, tools, or governance models have you adopted to maintain integrity and depth in AI-led research and insight creation? - Have you had success (or struggles) balancing speed vs. substance in AI-powered strategies? Your thoughts, examples, and recommendations would be incredibly valuable in helping others try to navigate this shift responsibly.

Does your company have a data recovery process in place?

Yes84%

No12%

Not sure3%

View Results