TCS
Large multinational IT services and consulting enterprise based in India.
3 Rounds
~14 Days
Medium
The Interview Loop
Recruiter Screen (30 min)
Standard fit check, behavioral questions, and resume overview.
Technical Loop (3-4 Rounds)
Deep dive into domain knowledge, coding, and system design.
Interview Question Bank
Machine Learning Engineer
•
Behavioral
•
medium
Tell me about a time you had to explain the results and limitations of a complex Machine Learning model to a non-technical client stakeholder.
#Stakeholder Management
#Storytelling
#Consulting
Machine Learning Engineer
•
Behavioral
•
medium
Describe a situation where a client provided you with data that was of extremely poor quality or lacked labels. How did you handle the project?
#Data Cleaning
#Client Management
#Ambiguity
Machine Learning Engineer
•
Behavioral
•
medium
At TCS, you might work on multiple Proof of Concepts (PoCs) simultaneously. How do you prioritize your tasks when facing tight, conflicting deadlines?
#Time Management
#Agile
#Prioritization
Machine Learning Engineer
•
Behavioral
•
hard
Tell me about a time a machine learning model you deployed failed or performed poorly in production. What was your Root Cause Analysis (RCA) process?
#RCA
#Production Issues
#Accountability
Machine Learning Engineer
•
Behavioral
•
easy
The AI/ML landscape is evolving rapidly (e.g., GenAI). How do you stay updated, and can you give an example of how you applied a recently learned concept to a project?
#Continuous Learning
#AI Trends
#Adaptability
Machine Learning Engineer
•
Coding
•
medium
Implement a custom F1-score calculation function from scratch using pure Python and NumPy, without using scikit-learn.
#Python
#NumPy
#Evaluation Metrics
Machine Learning Engineer
•
Coding
•
medium
Given an array of intervals where intervals[i] = [start_i, end_i], merge all overlapping intervals. This is often used in processing time-series data for sensor logs.
#Arrays
#Sorting
#Data Structures
Machine Learning Engineer
•
Coding
•
medium
Write a SQL query to find the top 3 highest-grossing products in each category from a 'sales' table. Assume a standard enterprise retail schema.
#Window Functions
#Aggregation
#Data Engineering
Machine Learning Engineer
•
Coding
•
medium
You have a pandas DataFrame with 10 million rows containing missing values and outliers. How do you efficiently clean this data without running out of memory?
#Pandas
#Memory Optimization
#Data Preprocessing
Machine Learning Engineer
•
Coding
•
medium
Given a string representing a sequence of user actions, find the length of the longest substring without repeating characters.
#Sliding Window
#Strings
#Hash Map
Machine Learning Engineer
•
System Design
•
hard
Design a Retrieval-Augmented Generation (RAG) system for a client who wants an internal chatbot to query their proprietary HR policy PDFs.
#RAG
#Vector Databases
#LLMs
#System Architecture
Machine Learning Engineer
•
System Design
•
hard
Design an end-to-end ML system for real-time credit card fraud detection. The system must process 5,000 transactions per second with sub-50ms latency.
#Real-time Processing
#Streaming
#Fraud Detection
#Architecture
Machine Learning Engineer
•
System Design
•
hard
Design a personalized product recommendation engine for an e-commerce client. How do you handle the 'cold start' problem for new users and new products?
#Recommendation Systems
#Collaborative Filtering
#Cold Start
Machine Learning Engineer
•
System Design
•
hard
Design a pipeline to automatically extract key entities (Vendor Name, Invoice Date, Total Amount) from scanned PDF invoices for an accounting client.
#OCR
#NLP
#Information Extraction
#Pipeline Design
Machine Learning Engineer
•
System Design
•
hard
Design a batch prediction pipeline that scores 50 million customer records every night to predict churn. The data resides in an AWS S3 data lake.
#Batch Processing
#Cloud Architecture
#PySpark
#AWS
Machine Learning Engineer
•
System Design
•
medium
Design a scalable architecture for a customer churn prediction model that needs to be retrained weekly with the latest transactional data.
#Continuous Training
#Pipeline Orchestration
#Classification
Machine Learning Engineer
•
Technical
•
medium
Explain the mathematical difference between L1 (Lasso) and L2 (Ridge) regularization. Why does L1 lead to sparsity?
#Regularization
#Mathematics
#Feature Selection
Machine Learning Engineer
•
Technical
•
medium
For a BFSI client, we are building a credit card fraud detection model where the fraud rate is 0.01%. How do you handle this severe class imbalance?
#Imbalanced Data
#Classification
#SMOTE
Machine Learning Engineer
•
Technical
•
medium
Explain the internal working of XGBoost. How does it differ from a Random Forest algorithm?
#Ensemble Methods
#Decision Trees
#Gradient Boosting
Machine Learning Engineer
•
Technical
•
hard
What is the 'Curse of Dimensionality'? Explain the mathematics behind Principal Component Analysis (PCA) and how it mitigates this issue.
#Dimensionality Reduction
#Linear Algebra
#PCA
Machine Learning Engineer
•
Technical
•
medium
We are clustering customer profiles for a retail client, but we have no ground truth labels. How do you evaluate the quality of your clustering algorithm?
#Unsupervised Learning
#Clustering
#Evaluation Metrics
Machine Learning Engineer
•
Technical
•
easy
Explain the concepts of Bagging and Boosting. Give one real-world example where you would prefer one over the other.
#Ensemble Methods
#Model Selection
Machine Learning Engineer
•
Technical
•
medium
A model deployed for a client 6 months ago is now degrading in performance. How do you detect data drift and concept drift, and how do you resolve it?
#Model Monitoring
#Data Drift
#Production ML
Machine Learning Engineer
•
Technical
•
medium
Walk me through your exact strategy for tuning hyperparameters for a Gradient Boosting model to prevent overfitting.
#Hyperparameter Tuning
#Optimization
#Overfitting
Machine Learning Engineer
•
Technical
•
hard
Explain the Self-Attention mechanism in Transformer architectures. How are the Query, Key, and Value matrices generated and used?
#Transformers
#NLP
#Attention Mechanism
Machine Learning Engineer
•
Technical
•
medium
When would you choose to fine-tune an open-source LLM (like Llama 3) versus using a RAG approach with a commercial API (like GPT-4)?
#LLMs
#Fine-tuning
#RAG
#Model Selection
Machine Learning Engineer
•
Technical
•
medium
Explain the vanishing and exploding gradient problems in standard RNNs. How do LSTM networks solve this mathematically?
#RNN
#LSTM
#Optimization
Machine Learning Engineer
•
Technical
•
hard
A manufacturing client wants to deploy a defect detection deep learning model on edge devices (Raspberry Pi). How do you optimize the model for inference speed and size?
#Edge AI
#Quantization
#Model Pruning
#Computer Vision
Machine Learning Engineer
•
Technical
•
medium
Describe the architecture of a Convolutional Neural Network (CNN) you would build for a multi-class image classification task. What is the purpose of pooling layers?
#CNN
#Image Classification
#Deep Learning
Machine Learning Engineer
•
Technical
•
hard
What is LoRA (Low-Rank Adaptation) and why has it become the standard for fine-tuning Large Language Models in enterprise settings?
#PEFT
#LoRA
#LLM Fine-tuning
Machine Learning Engineer
•
Technical
•
medium
You are processing 100-page legal contracts using an LLM, but you are hitting the context window token limit. What strategies do you use to handle this?
#NLP
#Context Windows
#Chunking
Machine Learning Engineer
•
Technical
•
medium
Explain your approach to containerizing a Python-based ML model using Docker and deploying it on a Kubernetes cluster.
#Docker
#Kubernetes
#Model Deployment
Machine Learning Engineer
•
Technical
•
medium
We want to replace an existing rule-based system with a new ML model. What is your strategy for A/B testing the new model in production safely?
#A/B Testing
#Experimentation
#Deployment Strategies
Machine Learning Engineer
•
Technical
•
medium
Explain the core components of an MLOps pipeline. How have you used tools like MLflow or Weights & Biases in your previous projects?
#CI/CD
#MLflow
#Experiment Tracking
#Model Registry
Machine Learning Engineer
•
Technical
•
medium
How do you ensure model fairness and mitigate bias in a machine learning model built to approve or reject bank loans?
#AI Ethics
#Bias Mitigation
#Explainable AI
Difficulty Radar
Based on recent AI-sourced data.
Meet Your Interviewers
The "Standard" Interviewer
Senior EngineerFocuses on core competencies, system constraints, and clear communication.
SimulateUnwritten Rules
Think Out Loud
Always explain your thought process before writing code or drawing architecture.