i
Mirafra
Technologies
Filter interviews by
I appeared for an interview in Jun 2025, where I was asked the following questions.
Guardrails ensure safe AI usage; I have experience with NLIs like Rasa and Dialogflow for building conversational interfaces.
Guardrails are safety measures that guide AI behavior, ensuring ethical and reliable outputs.
In NLI, I have used Rasa for building custom chatbots that understand user intents.
I have experience with Dialogflow for creating voice-activated applications.
Guardrails can include constraints on data us...
Quantization is the process of mapping a continuous range of values to a finite range, often used in digital signal processing.
Reduces the number of bits needed to represent data, e.g., converting 32-bit floats to 8-bit integers.
Commonly used in machine learning to optimize models for faster inference and lower memory usage.
In image processing, quantization can reduce the number of colors in an image, e.g., converting ...
Cosine similarity measures the cosine of the angle between two vectors, indicating their similarity in vector databases.
Cosine similarity ranges from -1 to 1, where 1 means identical, 0 means orthogonal, and -1 means opposite.
It is particularly useful in high-dimensional spaces, such as text data represented as word embeddings.
For example, in a recommendation system, cosine similarity can help find similar items based ...
I applied via Naukri.com and was interviewed in Nov 2022. There were 3 interview rounds.
Top trending discussions
I applied via Referral and was interviewed before Oct 2023. There was 1 interview round.
Adaboost is a machine learning algorithm that combines multiple weak learners to create a strong learner.
Adaboost stands for Adaptive Boosting.
It works by adjusting the weights of incorrectly classified instances so that subsequent weak learners focus more on them.
The final prediction is made by combining the predictions of all the weak learners, weighted by their accuracy.
Example: Adaboost is commonly used in face det...
Printing a binary tree in different orders
Use inorder traversal to print the binary tree in ascending order
Use preorder traversal to print the binary tree in root-left-right order
Use postorder traversal to print the binary tree in left-right-root order
I applied via Campus Placement and was interviewed in Jul 2024. There were 2 interview rounds.
Few standard dsa questions were asked.
I applied via Company Website and was interviewed before May 2021. There was 1 interview round.
I applied via Approached by Company and was interviewed before May 2021. There were 2 interview rounds.
Based on NLP
I applied via Approached by Company and was interviewed before May 2021. There were 2 interview rounds.
Related to NLP
I applied via Naukri.com and was interviewed before Jul 2021. There were 2 interview rounds.
LSTM is a type of recurrent neural network that can remember previous inputs. BiLSTM is a variant that processes input in both directions.
LSTM stands for Long Short-Term Memory
LSTM can remember long-term dependencies in data
BiLSTM processes input in both forward and backward directions
BiLSTM is useful for tasks such as named entity recognition and sentiment analysis
I applied via Job Fair and was interviewed before Jun 2021. There were 2 interview rounds.
Python
based on 3 interview experiences
Difficulty level
Duration
based on 6 reviews
Rating in categories
Senior Software Engineer
327
salaries
| ₹10.7 L/yr - ₹20 L/yr |
Member Technical Staff
154
salaries
| ₹15.7 L/yr - ₹28 L/yr |
Software Engineer
131
salaries
| ₹4.1 L/yr - ₹10 L/yr |
Software Engineer2
69
salaries
| ₹6 L/yr - ₹13.5 L/yr |
Design & Verification Engineer
65
salaries
| ₹12 L/yr - ₹41.2 L/yr |
Apar Industries
Molex
TDK India Private Limited
Micron Technology