Taro Logo

Software Engineer in Test Interview Experience - France

September 1, 2024
Negative ExperienceNo Offer

Process

  1. Preliminary interview with the technical recruiter.

  2. Discussion with two SDETs (technical interview).

  3. Take-home technical test (one week to complete it).

  4. Interview with the VP of Testing.

The first two steps went very well. The technical recruiter was great, very friendly, and had excellent communication.

The second interview with the two SDETs went well too; they were friendly, and we discussed a few technical aspects.

On the other hand, the technical test (done at home) was a real disappointment.

It turned out to be of a significantly lower quality than what one would expect from a cutting-edge company in data science. The exercise required resetting the database every time the API tests were executed—an incomprehensible and fundamentally flawed request from a technical standpoint. This approach causes obvious interference between tests in the event of parallel runs, a basic concurrency management problem that any testing professional should recognize. This is an obvious mistake that should not have ended up in a serious technical test.

After taking the time to explain in detail why this approach was problematic during an interview with a testing manager, it became clear that there was a real disconnect between what is expected of candidates and best practices in test engineering. Although the manager was attentive, they did not officially acknowledge the issues, despite the factual arguments (they mentioned a potential post-mortem, but well, we'll see). Indeed, even in professional environments, such as Dataiku for example, it is unthinkable to reset the database with every test, because that would disrupt the testing.

Another fundamental point I raised concerns data validation via API tests, which is not their role. This task falls under integration tests. The mistake here is that the exercise did not respect the testing pyramid, which results in an inefficient structure where validation is done at the wrong level. Ignoring this principle demonstrates a poor understanding of test design. (You can do these checks on the API side if you don't have integration tests, but criticizing candidates for respecting the pyramid makes no sense).

Finally, I noticed a certain inconsistency in technical skills when difficulties were encountered in running the ultra-simple project even though it has a ReadMe file and three commands to copy and paste. This is worrying, especially coming from a company supposed to be at the forefront of technology and specializing in data science.

In conclusion, this experience was disappointing. The time and effort required for this exercise do not match the quality of the feedback and the expectations. If the company is serious about its recruitment process, it is crucial that it takes this criticism into account and corrects the errors in its technical exercises. Otherwise, it risks missing out on qualified candidates and continuing to use tests that make no technical sense.

Questions

  • Testing pattern (page object)
  • Containerization
  • Test pyramid
This content was translated from French
View Original

Was this helpful?

Interview Statistics

The following metrics were computed from 1 interview experience for the Dataiku Software Engineer in Test role in France.

Success Rate

0%
Pass Rate

Dataiku's interview process for their Software Engineer in Test roles in France is extremely selective, failing the vast majority of engineers.

Experience Rating

Positive0%
Neutral0%
Negative100%

Candidates reported having very negative feelings for Dataiku's Software Engineer in Test interview process in France.

Dataiku Work Experiences