Infosys DSE/SP On-Campus Interview – Selection Chances?
I wanted to share my actual Infosys interview experience. The entire interview lasted around 1.5 hours, and honestly, it was quite different from what I expected.
I have solved 1 easy + 3 test cases from complex problem and 1 test case from the medium and got shortlisted for the interview.
Note: I have used AI to help structure and organize this post for readability. The interview experience, questions, and details mentioned here are based on what I personally experienced.
- Coding Round
As soon as I entered the cabin, they gave me a laptop and asked me to solve a coding problem. I chose the first problem and started working on it.
I was given around 30 minutes.
The problem was something like this:
Given an array:
\- Elements at even indices are called the high pair.
\- Elements at odd indices are called the low pair.
For the high pair, we had to select two elements such that their product is maximum.
For example:
"\[1, 2, 3\]"
Choosing "2" and "3" gives the maximum product:
"2 × 3 = 6"
Then we had to do the same for the low pair, but this time we had to select two elements whose product is minimum.
Negative numbers could also be present.
For example:
"\[-4, -5, 6\]"
For the low pair, selecting "6" and "-5" gives:
"6 × (-5) = -30"
which is the minimum product.
Finally, we had to calculate the distance between the two products, rather than doing normal subtraction.
For example, if:
"High pair product = 10"
"Low pair product = -2"
Then the answer should be:
"|10 - (-2)| = 12"
because the distance between 10 and -2 is 12.
I managed to pass 11/12 test cases.
The interviewer then asked me to try to optimize the solution further. While I was working on the optimization, he came over, looked at what I was doing, and asked if we could start the technical interview.
I said yes.
\---
- Technical Interview
This part was honestly quite different from what I expected.
Since my resume had mostly AI-related projects, he started directly with my projects and kept asking follow-up questions.
For example, he asked me how neural networks work internally.
While I was explaining that, he suddenly asked:
«"What happens if the nodes don't have bias?"»
I answered that and then he continued with another follow-up question.
This continued for around 10 questions, where almost every answer led to another question.
After that, I showed him the website where I had deployed one of my projects.
He started asking questions about that as well.
For example, he asked what I had used in the backend.
I said FastAPI.
Then he asked:
«"Why do you use "@app.post(...)" above functions in FastAPI?"»
I explained that these are decorators and that they are used to associate the function with a particular HTTP route/method.
By this point, around 50 minutes had already passed since the interview started.
\---
- Operating Systems
Then he moved to OS.
He asked me a question related to processes and threads.
He basically asked what happens before threads are created.
I initially started explaining what happens when threads communicate with the kernel, but he stopped me and said something along the lines of:
«"You're telling me what happens afterwards. I'm asking what happens before that."»
He then explained what he was looking for.
This was probably one of the questions where I wasn't really sure what the interviewer was expecting.
\---
- DBMS / Indexing
After that, he moved to DBMS and asked about indexing.
I explained the approaches I knew, including:
\- B-Tree
\- Binary search
Then he asked:
«"What are the disadvantages of indexing?"»
I mentioned the additional space requirement.
But he wasn't satisfied.
He told me that there was another major problem and asked me to think about it.
I couldn't come up with it.
Then he explained that when data changes and new records are inserted, the indexes may also need to be updated/maintained, which creates additional overhead and puts more stress on the database.
That was something I hadn't considered during the interview.
\---
Final Thoughts
After that, he asked if I had any questions for him.
By that point, the interview had gone on for around 1.5 hours.
Overall, the interview was much more follow-up heavy than I expected.
The interviewer didn't just ask isolated questions. He would take whatever answer I gave and keep digging deeper into it.
Especially with my AI projects, it felt like:
Question → Answer → Follow-up → Follow-up → Follow-up
So knowing just the surface-level explanation of something on your resume probably isn't enough. You should be prepared to explain why you used it, how it works internally, and what happens if something changes.
I have asked for feedback and he told me that I need to improve my Core CS fundamentals and he also told me that I have did projects which have good potential and he told me that the explanation could be somewhat better from what you have already told me. Like he said that it is somewhat hard for people to understand and told me to improve my explanation by practicing with friends.
Honestly i want to know if I will get selected for DSE?