‹ Back to previous section Course home All page URLs

Top 50 Most Asked Machine Learning Interview Questions

Q: Is this enough to clear an interview?

It is the best starting point, not the finish line. After these, practise the company pages for your target employers and the sheets for your weaker topics.

The questions that appear most often across company-wise sets — master these first.

Why start here

We count how often each question appears across the 20 company-wise Machine Learning practice sets on this site and show the 50 most common. They are the safest questions to master first.

Solved — top 50Start practicing

3 QUESTIONS FREESign up to unlock the full question bank and track your progress.Unlock every question and answer, then continue practising company-wise and package-wise.

3 questions free Unlock the rest in seconds

How do you choose a classification threshold that maximises F1?

Classification ModelsIn 10 company setsMedium

Show answer

from sklearn.metrics import precision_recall_curve import numpy as np proba = clf.predict_proba(X_val)[:, 1] p, r, t = precision_recall_curve(y_val, proba) f1 = 2 * p * r / (p + r + 1e-9) best = t[np.argmax(f1[:-1])] print(best) Choose it on validation data, not the test set. scikit-learn 1.5+ also offers TunedThresholdClassifierCV.

Your random forest model file is 2 GB and too slow for real-time scoring. What can you do?

Decision Trees and EnsemblesIn 10 company setsMedium

Show answer

Reduce size by limiting max_depth and min_samples_leaf, using fewer trees since gains flatten after a few hundred, and dropping weak features. Alternatively switch to a gradient boosting model like LightGBM, which usually gives similar accuracy with far smaller, faster models. Converting to ONNX or using a compiled predictor can also speed up inference. Measure accuracy and latency after each change.

Train LightGBM with early stopping.

Decision Trees and EnsemblesIn 10 company setsMedium

Show answer

import lightgbm as lgb m = lgb.LGBMClassifier(n_estimators=5000, learning_rate=0.05, num_leaves=31, subsample=0.8, subsample_freq=1, colsample_bytree=0.8) m.fit(X_train, y_train, eval_set=[(X_val, y_val)], eval_metric='auc', callbacks=[lgb.early_stopping(100)]) print(m.best_iteration_) The large n_estimators is only an upper limit; early stopping picks the real number.

Your house price model's residuals grow larger for expensive houses. What do you do?

Regression ModelsIn 9 company setsMedium

Show answer

That is heteroscedasticity, typical for price data that is right-skewed. I would model log(price) instead of price, which turns absolute errors into roughly percentage errors and stabilises variance. Then I would convert predictions back with exp and check residual plots again. Evaluating with RMSLE or MAPE also fits better when relative error matters more than absolute error.

The fraud team can only review 500 alerts a day. How do you set the threshold?

Classification ModelsIn 9 company setsMedium

Show answer

Instead of 0.5, rank transactions by predicted probability and pick the threshold that yields about 500 alerts per day on recent data. Then report precision and recall at that cut-off, or precision at k. I would monitor daily alert volume because drift can change it, and recompute the threshold periodically. If cost data exists, I would choose the threshold that maximises expected savings.

Your dataset has a city column with 2,000 unique values. How will you encode it?

Data Preprocessing and Feature EngineeringIn 8 company setsMedium

Show answer

One-hot would create 2,000 sparse columns, so I would first group rare cities below a frequency threshold into 'Other'. Then options are frequency encoding, out-of-fold target encoding, or mapping cities to useful attributes such as tier, state or population. For boosting models like CatBoost or LightGBM, native categorical handling also works well. I would compare options with cross-validation.

Detect anomalies with Isolation Forest.

Clustering and Dimensionality ReductionIn 8 company setsMedium

Show answer

from sklearn.ensemble import IsolationForest iso = IsolationForest(n_estimators=300, contamination=0.01, random_state=42) iso.fit(X_train) scores = -iso.score_samples(X_new) flags = iso.predict(X_new) == -1 Higher scores mean more anomalous. Tune contamination to match how many alerts reviewers can handle.

Build RFM features for customer segmentation in pandas.

Clustering and Dimensionality ReductionIn 8 company setsMedium

Show answer

snap = df['order_date'].max() + pd.Timedelta(days=1) rfm = df.groupby('customer_id').agg( recency=('order_date', lambda d: (snap - d.max()).days), frequency=('order_id', 'nunique'), monetary=('amount', 'sum')) rfm[['frequency', 'monetary']] = np.log1p(rfm[['frequency', 'monetary']]) Scale these columns before clustering.

A new model version performs badly right after release. How do you handle it?

MLOps and DeploymentIn 8 company setsMedium

Show answer

Roll back immediately to the previous version from the model registry, which should be a one-step action. Then investigate: compare the input features served with those used in validation, check for training-serving skew, version mismatches or a wrong artifact. Add the failure case to automated pre-release checks, and use canary or shadow releases next time so issues affect only a small share of traffic.

A demand forecast is consistently under-predicting during festivals. How would you fix it?

Regression ModelsIn 7 company setsMedium

Show answer

The model lacks festival information. I would add features for festival dates such as Diwali or regional holidays, days-to-festival, and past festival uplift for the same item. If the business cost of stock-outs is higher than over-stock, I would use a quantile loss for a higher percentile. Then validate specifically on past festival periods, not just the average error.

Random forest and LightGBM give the same AUC. Which would you deploy?

Decision Trees and EnsemblesIn 7 company setsMedium

Show answer

I would compare beyond AUC: model size, inference latency, training time, calibration and stability across time periods. LightGBM is usually smaller and faster to score, while random forest needs less tuning and is less prone to overfitting. If both are equal, I would choose the one that is cheaper to run and easier for the team to maintain and retrain.

Marketing wants customer segments for campaigns. How do you approach it?

Clustering and Dimensionality ReductionIn 7 company setsMedium

Show answer

I would agree on the purpose first, such as retention or upsell. Then build RFM features (recency, frequency, monetary) plus category preferences, log-transform skewed values and scale them. I would try k-means for k between 3 and 8, use silhouette and business review to pick k, profile each segment and give them clear names. Finally, test segment-based campaigns against a control group.

A new model has slightly better AUC but worse precision in the top 1,000 cases the team acts on. Which do you pick?

Model Evaluation and TuningIn 7 company setsMedium

Show answer

I would pick the model that does better where decisions are actually made. If the team only reviews the top 1,000 cases, precision at 1,000 or recall at that cut-off matters more than global AUC. I would make that the primary metric in future comparisons, and confirm the result holds across several time periods before switching.

Your manager asks for a single number showing the model's business value. What do you give?

Model Evaluation and TuningIn 7 company setsMedium

Show answer

I would convert model outcomes into money or time using a cost matrix agreed with the business, for example savings from caught fraud minus the cost of reviewing false alerts. I would compare this against the current process or a rule-based baseline on holdout data and show a range rather than one point. A short note on assumptions keeps the number honest.

Your team retrains models manually in notebooks every month. How would you improve it?

MLOps and DeploymentIn 7 company setsMedium

Show answer

I would convert the notebook into a versioned training pipeline with separate steps for data extraction, validation, feature building, training and evaluation, scheduled with Airflow or a cloud pipeline. Each run would log to MLflow, compare against the production model on a fixed holdout, and register the new model only if it passes thresholds. Promotion would stay a reviewed step, at least initially.

Log a training run to MLflow.

MLOps and DeploymentIn 7 company setsMedium

Show answer

import mlflow, mlflow.sklearn mlflow.set_experiment('churn') with mlflow.start_run(): pipe.fit(X_train, y_train) auc = roc_auc_score(y_val, pipe.predict_proba(X_val)[:, 1]) mlflow.log_params({'C': 1.0, 'model': 'logreg'}) mlflow.log_metric('val_auc', auc) mlflow.sklearn.log_model(pipe, 'model', registered_model_name='churn-model')

Use HistGradientBoosting with native categorical support in scikit-learn.

Decision Trees and EnsemblesIn 7 company setsMedium

Show answer

from sklearn.ensemble import HistGradientBoostingClassifier clf = HistGradientBoostingClassifier( categorical_features='from_dtype', max_iter=500, learning_rate=0.05, early_stopping=True, random_state=42) X_train[cat_cols] = X_train[cat_cols].astype('category') clf.fit(X_train, y_train) It handles missing values natively and is much faster than GradientBoostingClassifier on large data.

Leadership wants a model in one week, but data quality is poor. How do you handle it?

ML FundamentalsIn 6 company setsMedium

Show answer

I would agree on a narrow first version: a simple, explainable baseline built on the cleanest available fields, with clear notes on data gaps and expected accuracy. In parallel I would log specific data issues and their impact. Delivering a working baseline with an honest limitations section builds trust and gives a reference point for improvement in later iterations.

How do you make an ML experiment reproducible?

ML FundamentalsIn 6 company setsMedium

Show answer

Fix random seeds in NumPy, Python and the model, pin library versions in requirements.txt, version the data or record its hash, and save the exact code commit and hyperparameters. Tools like MLflow log parameters, metrics and artifacts per run. Use pipelines so preprocessing is identical each time, and keep the train-test split stored rather than recreated randomly.

40 percent of values in an important income column are missing. What do you do?

Data Preprocessing and Feature EngineeringIn 6 company setsMedium

Show answer

I would study the missingness pattern first, for example whether self-employed applicants skip it. I would add a binary was_missing feature, impute with the median within relevant groups or use IterativeImputer, and compare against models that handle missing values natively, like HistGradientBoosting or XGBoost. I would also raise the issue with the data owner so collection can improve.

How do you build a scikit-learn pipeline for mixed numeric and categorical data?

Data Preprocessing and Feature EngineeringIn 6 company setsMedium

Show answer

Use ColumnTransformer inside a Pipeline so preprocessing is fit only on training folds. from sklearn.compose import ColumnTransformer from sklearn.pipeline import Pipeline from sklearn.impute import SimpleImputer from sklearn.preprocessing import StandardScaler, OneHotEncoder num = Pipeline([('imp', SimpleImputer(strategy='median')), ('sc', StandardScaler())]) pre = ColumnTransformer([ ('num', num, num_cols), ('cat', OneHotEncoder(handle_unknown='ignore'), cat_cols) ]) model = Pipeline([('pre', pre), ('clf', LogisticRegression(max_iter=1000))])

How do you tune the alpha of Ridge regression using cross-validation?

Regression ModelsIn 6 company setsMedium

Show answer

RidgeCV tests many alphas efficiently; scale features first. from sklearn.linear_model import RidgeCV from sklearn.pipeline import make_pipeline from sklearn.preprocessing import StandardScaler import numpy as np m = make_pipeline(StandardScaler(), RidgeCV(alphas=np.logspace(-3, 3, 30))) m.fit(X_train, y_train) print(m[-1].alpha_) LassoCV and ElasticNetCV work the same way.

k-means gives one huge cluster and several tiny ones. What could be wrong?

Clustering and Dimensionality ReductionIn 6 company setsMedium

Show answer

Likely causes are unscaled features, where one large-range feature dominates distance; heavy skew or outliers pulling centroids; or data that is not naturally spherical. I would log-transform and scale features, remove or cap extreme outliers, and try different k values. If shapes are irregular, DBSCAN, HDBSCAN or a Gaussian Mixture Model may fit better.

Your dataset has 800 correlated features and the model trains slowly. What would you do?

Clustering and Dimensionality ReductionIn 6 company setsMedium

Show answer

I would remove constant and near-duplicate features, drop one from each highly correlated pair, then apply PCA inside the pipeline and keep components explaining about 95 percent of variance. Alternatively, use L1-regularised models or tree importances for supervised selection. I would compare speed and cross-validated scores; often 50 to 100 features give nearly the same accuracy.

Grid search with 5 hyperparameters is taking two days. How do you speed it up?

Model Evaluation and TuningIn 6 company setsMedium

Show answer

Switch to random search or Optuna with pruning, which find good settings with far fewer trials. Narrow ranges using a quick coarse search, tune the most important parameters first, use early stopping for boosting, run trials in parallel with n_jobs, and tune on a stratified sample of the data before a final refit on the full set. HalvingRandomSearchCV is another quick option.

Tune LightGBM with Optuna.

Model Evaluation and TuningIn 6 company setsHard

Show answer

import optuna from sklearn.model_selection import cross_val_score def objective(trial): p = {'num_leaves': trial.suggest_int('num_leaves', 15, 255), 'learning_rate': trial.suggest_float('learning_rate', 0.01, 0.2, log=True), 'min_child_samples': trial.suggest_int('min_child_samples', 10, 200)} m = lgb.LGBMClassifier(n_estimators=500, **p) return cross_val_score(m, X, y, cv=5, scoring='roc_auc').mean() study = optuna.create_study(direction='maximize') study.optimize(objective, n_trials=50)

A model's accuracy dropped 10 percent over three months in production. How do you investigate?

MLOps and DeploymentIn 6 company setsMedium

Show answer

I would first check data pipelines for broken features, nulls or schema changes, then compare current feature distributions with training data using PSI to find drift. I would check whether the label definition or business process changed, and analyse errors by segment. If drift is real, retrain on recent data, possibly with time-weighted samples, and set up automatic drift alerts and scheduled retraining.

Serve a scikit-learn model with FastAPI.

MLOps and DeploymentIn 6 company setsMedium

Show answer

from fastapi import FastAPI from pydantic import BaseModel import joblib, pandas as pd app = FastAPI() model = joblib.load('model_v1.joblib') class Item(BaseModel): age: int income: float city: str @app.post('/predict') def predict(item: Item): df = pd.DataFrame([item.model_dump()]) return {'score': float(model.predict_proba(df)[0, 1])} Run it with: uvicorn app:app --host 0.0.0.0 --port 8000

What is the curse of dimensionality?

ML FundamentalsIn 5 company setsMedium

Show answer

As the number of features grows, data becomes sparse in the feature space, distances between points become similar, and the amount of data needed to generalise grows rapidly. Distance-based methods like k-NN and k-means suffer most. Remedies include feature selection, dimensionality reduction such as PCA, regularisation and collecting more data. Text and genomics data often face this problem.

You are asked to predict customer churn. How would you frame the ML problem from scratch?

ML FundamentalsIn 5 company setsMedium

Show answer

First define churn precisely, for example no purchase in 60 days, and the prediction window. It becomes binary classification. Build features from data available before the cut-off date only, choose a time-based split, and pick metrics like recall at a fixed precision or lift in the top decile. Start with logistic regression as a baseline, then try boosting, and agree on how retention teams will use the scores.

How would you build a quick baseline before trying complex models?

ML FundamentalsIn 5 company setsEasy

Show answer

Use a DummyClassifier or DummyRegressor to see the score of a trivial guess, then a simple model. from sklearn.dummy import DummyClassifier base = DummyClassifier(strategy='most_frequent') base.fit(X_train, y_train) print(base.score(X_test, y_test)) Any real model must clearly beat this number; otherwise the features carry little signal.

How do you implement simple gradient descent for linear regression in NumPy?

ML FundamentalsIn 5 company setsMedium

Show answer

Compute predictions, the gradient of MSE, and update weights in a loop. import numpy as np w = np.zeros(X.shape[1]); b = 0.0; lr = 0.01 for _ in range(1000): err = X @ w + b - y w -= lr * (2 / len(y)) * (X.T @ err) b -= lr * (2 / len(y)) * err.sum() Scale features first or the loop converges slowly.

Tell me about a time your model did not work as expected. What did you do?

ML FundamentalsIn 5 company setsMedium

Show answer

Situation: my first churn model scored 0.95 AUC in validation but failed in a pilot. Task: find out why. Action: I traced features and found one field updated after customers left, which leaked the label. I removed it, rebuilt features with a strict cut-off date and moved to a time-based split. Result: AUC dropped to 0.78 but held steady in production, and I added a leakage checklist to my workflow.

Why do skewed features often need a log or power transform?

Data Preprocessing and Feature EngineeringIn 5 company setsMedium

Show answer

Heavily right-skewed features like income or transaction amount have long tails that can dominate linear models and distance metrics. A log1p, Box-Cox or Yeo-Johnson transform compresses large values and makes the distribution closer to normal, which often improves linear and distance-based models. Yeo-Johnson works with zero and negative values. Tree models usually do not need these transforms.

Validation AUC is 0.99 on a loan default model. What do you suspect?

Data Preprocessing and Feature EngineeringIn 5 company setsMedium

Show answer

Such a high score on a hard problem usually signals leakage. I would check feature importances for a suspiciously dominant feature, look for fields recorded after the loan outcome such as collection status, verify preprocessing was fit only on training data, and check for duplicate customers across train and test. Then re-run with a time-based split to confirm a realistic score.

How do you approach a time series forecasting problem differently from normal regression?

Regression ModelsIn 5 company setsMedium

Show answer

Time series data is ordered and autocorrelated, so random splits leak future information. I use time-based splits or TimeSeriesSplit, create lag and rolling-window features from past values only, and handle trend and seasonality. Baselines like last value or seasonal naive come first. Models range from ARIMA and Prophet to gradient boosting on lag features.

Two models have RMSE 120 and 125, but the second has much lower MAE. Which do you choose?

Regression ModelsIn 5 company setsMedium

Show answer

It depends on error cost. The first model makes fewer big errors, while the second is better on typical cases but has some large misses. If large errors are expensive, such as stock-outs of costly items, choose the first. If typical accuracy matters more, choose the second. I would also check error distributions by segment and discuss with stakeholders before deciding.

You need prediction intervals, not just point estimates, for delivery time. How?

Regression ModelsIn 5 company setsHard

Show answer

Options are quantile regression, training separate models for the 10th and 90th percentiles with GradientBoostingRegressor(loss='quantile') or LightGBM; conformal prediction, which wraps any model and gives intervals with guaranteed coverage; or bootstrapping. I would check calibration: about 80 percent of actual values should fall within an 80 percent interval on a holdout set.

How do you create lag and rolling features for a sales forecasting model in pandas?

Regression ModelsIn 5 company setsHard

Show answer

Sort by item and date, then shift so features use only past values. df = df.sort_values(['item', 'date']) g = df.groupby('item')['sales'] df['lag_7'] = g.shift(7) df['roll_mean_28'] = g.transform(lambda s: s.shift(1).rolling(28).mean()) The shift before rolling prevents the current day's sales from leaking into its own feature.

What is boosting and how is it different from bagging?

Decision Trees and EnsemblesIn 5 company setsMedium

Show answer

Boosting trains models sequentially, each one focusing on the errors of the combined previous models, which mainly reduces bias. Bagging trains models independently in parallel and averages them, which mainly reduces variance. Boosting uses shallow trees and a learning rate, and can overfit if run too long, so early stopping is important. Examples are AdaBoost, Gradient Boosting, XGBoost, LightGBM and CatBoost.

What makes XGBoost, LightGBM and CatBoost different?

Decision Trees and EnsemblesIn 5 company setsMedium

Show answer

XGBoost adds regularisation to the objective, uses second-order gradients and handles missing values natively. LightGBM grows trees leaf-wise and uses histogram binning, making it very fast on large data, though it can overfit small data. CatBoost handles categorical features with ordered target statistics and uses symmetric trees, giving strong defaults with less tuning. All three support GPU training.

How would you encode cyclical features like hour of day or month?

Data Preprocessing and Feature EngineeringIn 5 company setsMedium

Show answer

Map the value onto a circle with sine and cosine so the end connects to the start. import numpy as np df['hour_sin'] = np.sin(2 * np.pi * df['hour'] / 24) df['hour_cos'] = np.cos(2 * np.pi * df['hour'] / 24) Now 23:00 and 00:00 are close, which linear and distance-based models cannot see with raw integers. Tree models often manage with the raw value.

When should you use TimeSeriesSplit or GroupKFold instead of normal k-fold?

Model Evaluation and TuningIn 5 company setsMedium

Show answer

Use TimeSeriesSplit when data is ordered in time, so each validation fold comes after its training data and no future information leaks back. Use GroupKFold when multiple rows belong to the same entity, such as several visits per patient, so the same entity never appears in both train and validation. Normal k-fold in these cases gives over-optimistic scores.

What fairness checks would you run on a model that affects people?

Model Evaluation and TuningIn 5 company setsHard

Show answer

Compare metrics across groups such as gender, age band or region: approval rate (demographic parity), true positive rate (equal opportunity) and false positive rate. Check calibration per group too. Remove or audit proxy features like pin code that may stand in for protected attributes. Document findings in a model card, and involve legal or compliance teams, especially for lending and hiring use cases.

Plot a learning curve in scikit-learn.

Model Evaluation and TuningIn 5 company setsMedium

Show answer

from sklearn.model_selection import LearningCurveDisplay import numpy as np LearningCurveDisplay.from_estimator(model, X, y, cv=5, train_sizes=np.linspace(0.1, 1.0, 6), scoring='roc_auc', n_jobs=-1) plt.show() A persistent gap between curves suggests more data or regularisation; two low curves suggest a richer model or better features.

Get a bootstrap confidence interval for a test metric.

Model Evaluation and TuningIn 5 company setsHard

Show answer

import numpy as np from sklearn.metrics import roc_auc_score rng = np.random.default_rng(0); scores = [] yt, pp = np.array(y_test), np.array(proba) for _ in range(1000): i = rng.integers(0, len(yt), len(yt)) if len(np.unique(yt[i])) == 2: scores.append(roc_auc_score(yt[i], pp[i])) print(np.percentile(scores, [2.5, 97.5]))

Compare batch scoring and real-time inference.

MLOps and DeploymentIn 5 company setsMedium

Show answer

Batch scoring runs on a schedule, for example nightly churn scores for all customers written to a table; it is simple, cheap and tolerant of slow models. Real-time inference returns predictions on request, such as fraud checks during payment, and needs low latency, high availability, scaling and fresh features. Choose real-time only when the decision must happen at the moment of the event.

Labels for your loan model arrive only after 6 months. How do you monitor it meanwhile?

MLOps and DeploymentIn 5 company setsHard

Show answer

I would rely on proxy signals: input data drift with PSI, prediction score distribution, approval rates by segment, and early indicators like first missed payment at 30 or 60 days. Once enough early outcomes arrive, I would check performance on these. I would also set up a delayed evaluation job that computes full metrics as labels mature and compares them with the validation baseline.

Write a unit test for a model prediction function.

MLOps and DeploymentIn 5 company setsMedium

Show answer

import joblib, pandas as pd def test_predict_shape_and_range(): model = joblib.load('model_v1.joblib') df = pd.DataFrame([{'age': 30, 'income': 50000.0, 'city': 'Pune'}, {'age': 45, 'income': 0.0, 'city': 'UnknownCity'}]) p = model.predict_proba(df)[:, 1] assert p.shape == (2,) assert ((p >= 0) & (p <= 1)).all() Run with pytest in CI so broken artifacts never reach production.

A product team wants 'similar products' recommendations with no purchase labels. How would you start?

Clustering and Dimensionality ReductionIn 5 company setsMedium

Show answer

I would build product embeddings from content: TF-IDF or sentence embeddings of titles and descriptions, plus category and price features. Then use cosine similarity with nearest-neighbour search, such as scikit-learn NearestNeighbors or an approximate index like FAISS for large catalogues. I would review samples with the product team and later add co-view or co-purchase signals once collected.

Frequently asked questions

How is this list made?

We count how often each question appears across the 20 company-wise Machine Learning practice sets on this site and show the 50 most common. They are the safest questions to master first.

Is this enough to clear an interview?

It is the best starting point, not the finish line. After these, practise the company pages for your target employers and the sheets for your weaker topics.

More ways to practise

Company-wise Machine Learning questions Experience-wise Machine Learning questions Role-wise Machine Learning questions City-wise Machine Learning questions Scenario-based questions Practical / hands-on questions HR & behavioural questions Machine Learning mock interview Online assessment questions Technical round questions Managerial round questions Machine Learning topic sheets