applied in January, got the offer in April, documenting everything because I wish I had this before I started.
the new grad (MTS 1 equivalent, or just 'software engineer' in the JD) loop at MongoDB is not a grind-150-leets situation but it is not casual either. here's what I actually saw:
phone screen (45 min): one interviewer, LC-style question, medium difficulty. I got a trees problem. clean solution expected but not necessarily the most optimized one. they want to see you talk through your thinking. running through edge cases out loud helped.
onsite (4 rounds via Zoom, all same day): coding 1: medium/medium-hard. arrays or graphs, I don't want to be too specific. finish in 35-40 min with time to explain. coding 2: design-lite, more like 'how would you structure this?' not a full system design. I think for new grads they're testing whether you can reason about code organization, not Kafka vs RabbitMQ. behavioral: pure STAR. they asked about a team conflict, a time I shipped something imperfect, and a time I disagreed with a technical direction. write these stories BEFORE the interview, not at 11pm the night before (learned this the hard way for an earlier company). hiring manager / cultural fit: more of a conversation, asked about where I want to be in 3 years, what I know about MongoDB's products. I had done homework on Atlas and the vector search stuff and it showed.
for prep: LC medium is the right level, focus on graph traversal, trees, two pointers, and dynamic programming basics. MongoDB is a database company so they want to see you're comfortable with data-structure thinking.
comp for NYC new grad came out around $165-175k total first year based on what I heard from a friend in the same batch (I was in a different office). base is solid, signing bonus was $20k, RSUs vest over 4 years.
also: the recruiter was very responsive. like, replied within a few hours. that alone was a green flag after months of ghosting elsewhere.