Recent US reports describe the same Databricks interview process: a 20-30 minute recruiter call, a 60-minute technical screen with one LeetCode-style problem you must also test, and a virtual onsite of four interviews covering algorithms, practical coding or API design, a systems or system design round, and a hiring manager behavioral. Candidates who got offers called the questions fair and not obscure; those who were rejected mostly fell at system design or ran short of time on follow-ups.
Databricks interview process round by round
| Round | Duration | What candidates reported |
|---|---|---|
| Recruiter call | ~20 min | Resume and internship walkthrough, including how the internship ended |
| Technical screen | 60 min | One problem such as Design Hit Counter, plus your own test cases |
| Onsite: algorithms | ~60 min | LeetCode medium; a bit manipulation question in one 2025 loop |
| Onsite: coding or API | ~60 min | An implementation problem or “construct an API for a practical problem” |
| Onsite: systems or design | ~60 min | Systems round (concurrency, memory) or architecture design, depending on the loop |
| Onsite: behavioral | ~45-60 min | Past projects and internships, fit with Databricks |
Onsite durations are not stated in the reports; they are listed here by analogy with the 60-minute screen.
Databricks interview process in 2025: what candidates reported
These are self-reported 2025 accounts on Taro (jointaro.com). Each is one person’s experience.
- Software engineer, United States (November 2025, accepted offer). One technical screen, then an onsite of four rounds: DSA, systems (the candidate stressed it was not system design), coding and behavioral. The only question they shared was a bit manipulation problem that was not obscure. They said the recruiter and hiring manager were very responsive and helped them prepare.
- Software engineer, United States (October 2025, offer). One LeetCode-style problem in the tech screen, then a virtual onsite of a LeetCode problem, a coding or implementation problem, an architecture or system design round and a behavioral round where they were asked about past projects and internships. Everything felt manageable except system design, which they had little experience with as a new grad.
- Software engineer, United States (October 28, 2025, no offer). A one-hour phone screen. After brief introductions the interviewer went straight to the hit counter design problem and asked for test cases. The candidate rated the experience positive.
- New grad software engineer, San Francisco (October 2025, no offer). Recruiter call, a first-round interview, then a final round of four interviews including system design. One task was to construct an API for a practical problem. They called the process very difficult and the interviewers very nice.
- New grad software engineer, Bellevue, WA (September 2025, no offer). A 20-minute recruiter call covering their background, the role they wanted, and what they did at their previous internship and how it ended, then a 60-minute technical interview with an engineer.
- Software engineering intern, Mountain View, CA (October 2025, no offer). Two LeetCode-style medium coding interviews, each with follow-ups, then a purely behavioral hiring manager round about past experience and fit. Their advice was to practice Databricks-tagged problems.
Detailed interview experiences
Experience 1: Software engineer, US, offer accepted (November 2025)
Screen: One technical phone screen; the candidate did not share the problem.
Onsite: Four rounds. The DSA round included a bit manipulation question that the candidate said was not obscure or intimidating. The systems round was explicitly not a system design interview, so prepare concurrency, memory and data-structure trade-offs rather than only boxes and arrows. A separate coding round and a behavioral round followed.
What helped: The recruiter and hiring manager were proactive in helping the candidate prepare, so ask your recruiter what each round covers.
Experience 2: New grad, US, offer (October 2025)
Screen: One LeetCode-style problem.
Onsite: A LeetCode problem, an implementation problem, an architecture or system design round, and a behavioral round on past projects and internships.
Where it was hard: System design. As a new grad the candidate had little practice with it. If you are a new grad, do not assume Databricks will skip design; at least two 2025 new grad loops included it.
Experience 3: New grad, San Francisco, reached final round (October 2025)
Process: Recruiter call, a first-round interview, then a four-interview final round with a system design component.
Question shared: Construct an API for a practical problem.
Result: No offer. The candidate found the process very difficult but liked the interviewers.
Takeaway: Practice designing small APIs end to end: endpoints, data model, error cases, and how the API behaves under concurrent calls.
Experience 4: Phone screen with Design Hit Counter (October 2025)
Format: One hour. Brief introductions, then live coding.
Problem: Record hits with timestamps and return how many arrived in the past five minutes, using a sliding window. The candidate had to implement it and write test cases.
Result: No offer, though they rated the experience positive.
Takeaway: Write tests without being asked, and know both the queue solution and the fixed 300-bucket solution with their memory trade-offs.
Experience 5: Intern, Mountain View (October 2025)
Rounds: Two LeetCode medium interviews with follow-ups, then a behavioral hiring manager interview about past experience and fit with Databricks.
Result: No offer. The candidate’s advice was to practice Databricks-tagged problems.
Databricks interview questions that come up often
Coding
- Design Hit Counter (sliding window, five-minute count)
- A bit manipulation problem
- Construct an API for a practical problem
- Databricks-tagged practice: Design Tic-Tac-Toe, IP to CIDR, Task Scheduler, Web Crawler Multithreaded
Systems and design
- Concurrency and memory trade-offs in a systems round
- An architecture or system design round, including for some new grads
Behavioral
- Walk me through your internship and how it ended
- Tell me about your past projects
- Why Databricks?
Sample answers for the coding problems are on the Databricks interview questions hub, and behavioral answers mapped to the culture principles are on the Databricks behavioral interview questions page.
Databricks interview tips from these reports
Technical screen
- Expect one problem in an hour and test cases written by you.
- Expect follow-ups once the first version works.
Onsite
- Ask whether your loop has a systems round or a system design round; they are different.
- New grads in two 2025 loops still had system design.
Behavioral
- Hiring managers go deep on internships and past projects.
- Tie stories to the six culture principles with numbers.
Similar companies
Snowflake · Datadog · Cloudflare · Coinbase · ServiceNow · Uber
Sources
- Taro: software engineer, United States, November 2025 (accepted offer)
- Taro: software engineer, United States, October 2025 (offer)
- Taro: software engineer, United States, October 28, 2025
- Taro: new grad, San Francisco, October 2025
- Taro: new grad, Bellevue, September 2025
- Taro: intern, Mountain View, October 2025
- AlgoMonster: Databricks-tagged problem list