Jupiter

Role: ML Engineer

College: Not Specified

Interview Date: April 2022

Interview Type: Off-Campus

Result: Selected

Difficulty: Medium to Hard

Rounds: 3 rounds

Topics Asked: Data Structures, Algorithms, Graph Theory, Technical MCQs, Behavioral, Deep Learning, Operating Systems, Networking, Asynchronous Code, Parallel Programming

Experience:

The interview process for the ML Engineer role at Jupiter Money consisted of three rounds.

Round 1: Coding/Problem-Solving Round (90 minutes)
This round was conducted on the CoderByte platform and included 2 GFG hard/expert level graph questions, 13 technical MCQs, and 2 optional behavioral questions. One graph problem was a variation of finding the longest path between two nodes (diameter of a graph), and the second was a variation of topological sorting. The candidate successfully solved both coding questions.

Round 2: Technical Interview (Online - 75 minutes)
After clearing Round 1, the candidate was informed of their selection for this round on the same day, with the interview scheduled three days later. The interviewer was friendly and started with introductions. The round also took place on CoderByte. The interviewer asked the candidate to explain their summer internship project (deep learning) in detail, which led to a discussion. Following this, DSA questions were asked, which were described as GFG easy/medium. Questions included:

• Wave Sort an array (optimize to O(N) solution).
• Find the most frequently occurring word in a resume (clarified as .txt format, suggested hash map approach, discussed hash collision mitigating techniques like Open addressing, Chaining).
• Asynchronous code and parallel programming techniques.
• Write an SQL query to find the Nth highest salary of an employee from the salary table.
• Peak finding variation (Binary Search) and coding the solution.

The round concluded with the candidate asking questions about the role and expectations.

Round 3: Hiring Manager Round (Online - 35 minutes)
This round was scheduled four days after Round 2 with the Data Science project team manager. It began with introductions and then the manager inquired about the previous rounds and the candidate's interview experience. The discussion then shifted to the candidate's Machine Learning project, including its features and the choice of technology, with follow-up questions. Operating system concepts like paging and memory management were also covered. The interviewer asked for an explanation of the entire flow of a request in the network OSI model when hitting a URL. Finally, generic questions like career aspirations (2-3 years) and expectations were asked.

Tips: Be confident with projects, clarify anything not understood, and it's acceptable not to know everything.