E source Data Engineer interview typically runs 2 rounds: a 7-day take-home assignment and a 90-minute presentation. The process takes about a week and is unusually heavy for a Data Engineer role.
$120K
Avg. Base Comp
$195K
Avg. Total Comp
2
Typical Rounds
1-2 weeks
Process Length
Our candidates report that E Source cares less about polished interview banter and more about whether you can independently produce something that looks like client-ready engineering work. The standout signal is the depth of the take-home: multiple artifacts, multiple datasets, pipeline work, modeling, and a DevOps architecture narrative all bundled together. That tells us they are screening for people who can operate with a lot of ambiguity and still deliver a coherent, end-to-end solution in a consulting context.
What makes this process different is that the live conversation seems to function mainly as a defense of your decisions, not a discovery exercise. The candidate experience we saw was centered on explaining why the architecture was built a certain way, how the pipelines fit together, and why the modeling choices made sense. In other words, they appear to value decision quality and technical justification as much as the implementation itself. If your work is strong but your reasoning is fuzzy, that’s where you’ll likely lose them.
A recurring theme is the sheer amount of upfront effort relative to the role. The candidate who shared this experience described the assignment as closer to a full work week than a screening, and the lack of feedback afterward suggests the bar is high but the process can feel transactional. For us, that means E Source is probably best suited to candidates who are comfortable treating the interview like a mini consulting engagement and can present their work with crisp, defensible architecture choices.
Synthetized from 1 candidates reports by our editorial team.
Had an interview recently?
Share your experience. Unlock the full guide.
Real interview reports from people who went through the E source process.
The most frustrating part of this process was that the “interview” was really a 7-day take-home assignment that felt closer to a full work week than a screening. I had to set up Databricks and GitHub, work through multiple datasets, build data pipelines and workflows, and then do final data modeling before putting together a detailed DevOps architecture presentation. On top of that, they wanted all of the code, data pipeline artifacts, and slides submitted ahead of time, which made the whole thing feel unusually heavy for a Data Engineer role.
After I turned everything in, I was asked to present the work in a 90-minute session. That presentation was the only live part of the process, and it was very much centered on walking through what I built and explaining the architecture decisions behind it. What stood out to me was the lack of feedback afterward. I eventually got a generic rejection email, with no real debrief or constructive comments, which was disappointing given how much time and effort the assignment required. If you’re preparing for this company, expect a long take-home with a strong emphasis on Databricks, pipeline design, and being able to defend your data modeling and DevOps choices clearly in presentation format.
Prep tip from this candidate
Be ready for a long take-home that includes Databricks setup, pipeline/workflow building, and data modeling, then practice presenting your architecture clearly for a 90-minute review. I’d also prepare to hand over polished code and slides as part of the submission, since that was a core part of the process.
Share your own interview experience to unlock all reports, or subscribe for full access.
Sourced from candidate reports and verified by our team.
Topics based on recent interview experiences.
Featured question at E source
Transform a dataframe containing a list of user IDs and their full names into one that contains only the user ids and the first name of each user.
| Question | |
|---|---|
| Why Do You Want to Work With Us | |
| Your Strengths and Weaknesses | |
| Empty Neighborhoods | |
| 2nd Highest Salary | |
| Comments Histogram | |
| Employee Salaries | |
| Closest SAT Scores | |
| Top Three Salaries | |
| Subscription Overlap | |
| Merge Sorted Lists | |
| Cumulative Distribution | |
| Experiment Validity | |
| Download Facts | |
| SELECTive Wine Connoisseur | |
| Liked Pages | |
| Rolling Bank Transactions | |
| Customer Orders | |
| Average Quantity | |
| String Shift | |
| Last Transaction | |
| Random SQL Sample | |
| Manager Team Sizes | |
| Search Ratings | |
| Like Tracker | |
| Daily Logins | |
| Month Over Month | |
| Flight Records | |
| Prime to N | |
| Alphabet Sum |
Synthesized from candidate reports. Individual experiences may vary.
Candidates complete a substantial take-home project rather than a traditional early screening. The assignment involves setting up Databricks and GitHub, working across multiple datasets, building data pipelines and workflows, and finishing with data modeling and a DevOps architecture presentation. All code, pipeline artifacts, and slides must be submitted before the live presentation.
After submitting the take-home, candidates present their work in a live 90-minute session. The discussion focuses on walking through the solution, explaining pipeline design, and defending architecture, data modeling, and DevOps decisions.
The process ends with a decision communicated afterward, which in this experience was a generic rejection email. No detailed feedback or debrief was provided.