Grammarly ML Engineer candidates report a recruiter-led loop with technical screens, a research or project presentation, ML coding, ML system design and behavioral rounds across roughly six to seven stages.
$381K
Avg. Base Comp
$506K
Avg. Total Comp
6-7 rounds
Typical Rounds
Not reported
Process Length
Candidates who interviewed for Grammarly's ML Engineer role describe a loop built around real ML work more than puzzle-solving. Two full-time candidates report a research or past-project presentation, where interviewers asked about their thinking, the challenges they hit and what they learned. One of them contrasted this with whiteboard-style algorithm questions. Prepare one or two projects you can explain in depth, including trade-offs and what you would change.
The loop structure varies between reports. One candidate describes two phone screens followed by four to five virtual onsite rounds. Another describes a technical screen mixing conceptual ML questions with coding, then five virtual onsite rounds. Those rounds included technical interviews, a hiring manager conversation and behavioral interviews. The second candidate listed ML-focused coding, ML system design and behavioral questions as the technical content, and said one round weighted practical day-to-day coding over abstract problem solving. A third candidate mentioned ML theory questions and a few standard coding problems.
Recruiters get consistent praise. Two candidates say the recruiter explained each round or set up extra team conversations, so ask for a round-by-round breakdown early. Three of the four reports ended in a declined offer.
The fourth report, from an ML engineering intern, describes a different process: an online assessment with four LeetCode-style problems, OOP and optimization questions, and resume deep-dives. That candidate waited almost two months and received no offer. Treat it as intern-level context rather than the full-time loop. Even so, it suggests being ready to explain failures and setbacks as well as wins.
Synthesized from 4 candidate 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 Grammarly process.
I applied to Grammarly in late October and got the OA a couple weeks later in mid-November. The coding portion was challenging—four LeetCode-style problems covering string manipulation and combinatorics. These weren't trivial; they required solid algorithmic thinking and clean implementation under time pressure. I also remember one problem specifically testing optimization, like converting a linear search to binary search within a class method, which meant understanding both the algorithm and OOP principles. Then there were questions around stack data structures and character handling that pushed me to think carefully about edge cases.
After clearing the coding round, I moved to the phone interviews. The interviewers were thorough but I felt less confident here than I'd hoped. We dove into my resume experiences in depth—what I liked most about a project, what I liked least, my biggest achievement, and my biggest challenge. Each answer got follow-ups, and they really wanted to understand my reasoning and growth mindset. They asked about career goals, what I thought of Grammarly's product, and gave me time for questions.
The whole process dragged on. There was significant waiting between rounds—almost two months total, with a particularly long gap after my interviews before I heard back. Communication was sparse, and I didn't receive any feedback on my performance. When the rejection finally came, there was no explanation. That silence and the timeline were frustrating, especially since I'd invested real effort into preparing and the interviews felt substantive enough that I thought I had a shot.
Prep tip from this candidate
Brush up on OOP design and optimization trade-offs—they'll ask you to implement a class hierarchy or system design and then defend your choices and refactor for complexity. Have concrete project stories ready for the resume deep-dive, but frame them around what you learned from failures and setbacks, not just wins.
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 Grammarly
Find the Length of the Largest Palindrome in a String
| Question | |
|---|---|
| Merge Sorted Lists | |
| Subscription Overlap | |
| First to Six | |
| Find Bigrams | |
| String Shift | |
| Bagging vs Boosting | |
| Find the Missing Number | |
| Assumptions of Linear Regression | |
| Scrambled Tickets | |
| Fair Coin | |
| Get Top N Frequent Words | |
| Job Recommendation | |
| P-value to a Layman | |
| 500 Cards | |
| The Brackets Problem | |
| Level Of Rain Water In 2D Terrain | |
| Raining in Seattle | |
| Type-ahead Search | |
| Payment Data Pipeline | |
| Youtube Recommendations | |
| Duplicate Rows | |
| Permutation Palindrome | |
| Lazy Raters | |
| Hurdles In Data Projects | |
| Decision Tree Evaluation | |
| Normal Distribution Sample | |
| Same Algorithm Different Success | |
| Nearest Common Ancestor | |
| Minimum Change |
Synthesized from candidate reports. Individual experiences may vary.
Candidates report starting with a recruiter call or conference contact that covered mutual fit before anything technical. Two candidates say the recruiter explained what each round would involve or arranged extra team conversations. It is reasonable to ask for a round-by-round breakdown at this stage.
Reports differ here. One candidate describes two relatively low-pressure phone screens establishing baseline technical fit. Another describes a single technical screen mixing conceptual ML questions with coding. You may see either shape, so prepare for both ML fundamentals and coding.
Two full-time candidates report presenting past research or projects. Interviewers asked follow-ups on the thinking, challenges and lessons behind the work, and one candidate contrasted this with algorithm-style questions. Prepare to discuss decisions, trade-offs and impact rather than only results.
Candidates report four to five or five virtual onsite rounds. One lists ML-focused coding, ML system design and behavioral questions, with one round favoring practical day-to-day coding. Another mentions ML theory questions and a few standard coding problems. Exact content likely varies by team.
Onsite loops reportedly included a hiring manager conversation and behavioral interviews. Interviewers were described as collaborative and knowledgeable. The intern report also describes resume deep-dives on favorite and least favorite projects, biggest challenge and career goals, so prepare concrete stories with specific lessons.
