Tip 1: Focus on understanding your projects deeply and be ready to explain the problem, approach, and challenges clearly.
Tip 2: Practice medium to hard-level coding questions, especially those involving graphs, OOP, and system design patterns.
Tip 3: Revise core CS fundamentals like OOP, DBMS, and Operating Systems-they’re often asked in interviews.
Tip 1: Highlight project outcomes and impact, not just the tools used-mention accuracy, performance improvements, or real-world relevance.
Tip 2: Keep it clean and concise-use bullet points, consistent formatting, and avoid clutter. Recruiters spend only a few seconds scanning it.
The environment was calm and professional. Since it was a virtual interview, I ensured a quiet setup with proper internet and no distractions. The interviewer maintained a friendly and conversational tone throughout, which helped ease any initial nervousness.
There weren’t any surprise activities, but the discussion was quite deep and project-focused. The interviewer gave me time to explain my projects in detail, especially the logic and challenges behind them. Towards the end, we also had a brief conversation about future goals and learning interests, which made the interview feel more personal and holistic.
You're navigating a grid of size N x M. Each cell has a positive value that represents the cost to step into it. You start from the top-left corner of the grid and need to reach the bottom-right corner.
Movement is allowed only in two directions—right and down. However, a few special cells have access to hidden shortcuts that let you jump instantly to another location in the grid. These shortcuts are rare and limited—you can't use too many.
Tip 1: Track teleport uses separately using a 3D visited array
Tip 2: Treat teleport links as 0-cost edges in your graph.
Tip 3: Avoid recomputing paths on every query—use smart caching if possible.
You have several projects, each with a duration, deadline, and reward (earned only if finished on time). You can do one project at a time without breaks, and you have a few helpers who can each reduce a project’s duration by half once. Choose which projects to do, in what order, and where to use your helpers to maximize your total reward before deadlines.
Tip 1:Sort projects by deadline to prioritize early completions.
Tip 2:Use 3D DP (project index, current time, helpers used) to track max rewards with/without helper usage.
Tip 3:At each step, choose to skip, do normally, or use a helper—update DP only if project finishes before its deadline.
Given a string s, you can repeatedly remove any palindromic subsequence (not necessarily contiguous). Each removal gives you points equal to the square of the length of the subsequence removed. After each removal, the remaining characters shift left to form a new string. Determine the maximum total score achievable by removing subsequences in an optimal order.
Tip 1: Use dp[l][r] to store max score for substring s[l...r].
Tip 2: Try removing all palindromic subsequences s[i...j] within s[l...r].
Tip 3: Cache results to avoid redundant computations and speed up recursion.
The interview was scheduled in the afternoon and started on time. It lasted for around 1 hour 45 minutes.
Nothing unusual happened during the interview, but the discussion was very project and research oriented. The interviewer took genuine interest in my machine learning work, especially the CNN project and my published paper. Toward the end, we also briefly discussed my future plans and areas I want to grow in.
The interviewer was polite, attentive, and encouraging. They asked insightful follow-up questions, gave me time to think, and kept the conversation engaging throughout. Their balanced and supportive approach really helped me stay confident during the entire session.
You are asked to design a basic machine learning pipeline structure using object-oriented programming principles. The system should support plugging in different models (like Random Forest, SVM, etc.) and allow easy training, evaluation, and prediction. The pipeline should also be extendable to support hyperparameter tuning in the future.
Tip 1: Start with a base abstract class Model that defines methods like train() and predict()
Tip 2: Follow SOLID principles: Especially Open/Closed Principle – your design should be open for extension (new models) but closed for modification.
Tip 3: Let your pipeline/controller class handle orchestration (training, evaluation), not model logic.



Tip 1: Use DFS or BFS to traverse connected cells.
Tip 2: Increment the island count after each full DFS traversal.
Tip 3: Be careful with edge conditions—check grid bounds during recursion.
Given a table Employee(salary INT), write a SQL query to find the 5th highest salary in the table. If there are fewer than 5 unique salaries, the query should return NULL or an empty result.
Tip 1: Use DISTINCT, ORDER BY, and LIMIT with OFFSET
Tip 2: Always check for duplicates—DENSE_RANK() is better when multiple employees can have the same salary.
Tip 3: For interviews, explain both methods and their trade-offs (especially if the interviewer asks about SQL window functions).

Here's your problem of the day
Solving this problem will increase your chance to get selected in this company
What is recursion?