Interview Questions

Sumit P. - "CQs: Content moderation system finds Inappropriate contents - profanity, violence, privacy concerning. Misinformation - false info, false claim, fomenting wrong views PII Misinformation → Wrong info Twisted info Incomplete info Goal - reliability and trust on the platform Long term increased engagement on informational content RAG system - what and why? RAG system has 3 components Brain - reasoning models Tool"See full answer

Christopher W. - "Stripe Connect is the backbone for marketplaces and platforms like think Shopify, DoorDash, or Lyft, that need to send money to multiple sellers or service providers. Its core promise is simple but vital: enable platforms to pay their users quickly, reliably, and compliantly. If we step back, a good North Star Metric should reflect the real value the product delivers to its users, and it should scale as that value grows. For Connect, that value is the flow of money from platforms to their conne"See full answer

Richa M. - "Before starting the answer I would align on goal, call out the structure where I cover what all am i going to talk about in the interview. Why laying out structure is important because the aprox time available to you is 35 min, so Structure shows the intent and the depth at which you are willing to go and think IMO This is how I would start- Sure before jumping in, I’ll first align on what this product is trying to achieve. Once we’re clear on the goal, I’ll define the key user segments, then"See full answer

Eshaa D. - "I started by reframing Adobe’s mission in my own words: to build creative and productivity tools that make everyday customer workflows easier. Then, I defined the goal: improving retention for the Scan app, as users often drop off to competitors such as Genius Scan or the native iOS Scanner after initial use. To better understand user behavior, I segmented users into three groups: Power users: Professionals such as lawyers or students who scan frequently. Low-engagement users: Inactive for 9"See full answer

Alex N. - "Since question asks about pipeline. I assume the question is about metrics across many dimensions not just prediction Model performance. For the ML Model: I can use accuracy, precision, recall, F1 if it is classification model. In case it is regression model RMSE is good metric for many problems. Data: ML system needs good quality data. The system has to track missing data rate. Distribution of features, if there is no drift from original feature distributions during the training. Pipeline h"See full answer

Alex N. - "The adjusting context window size in LLM change trade off between reasoning capability, accuracy, computation cost. It influence how attention mechanist allocate resources across the input. Longer context window let it you input greater number of words and have more context to generate proper next token. However llms have lost in the middle issue. They remember the beginning of text and end of text but lost information located in the middle of long input. Another problem is Attention Dilution."See full answer

Anonymous Partridge - "We will have to use a second more powerful LLM Model to validate the answers. LLM as a judge"See full answer

Hardik saurabh G. - "Hallucinations are evaluated by measuring how often generated outputs contain information that is not supported by trusted sources. what hallucination means in context: Intrinsic hallucination: contradicts provided context Extrinsic hallucination: introduces unsupported facts Fabrication: confidently incorrect answers"See full answer