PracHub
QuestionsPremiumLearningGuidesCheatsheetNEWCoaches
|Home/Machine Learning/Two Sigma

Forecast bikes available at a station

Last updated: Mar 29, 2026

Quick Overview

This question evaluates proficiency in time-series forecasting, feature engineering, handling temporal data splits, and incorporating operational constraints for predicting bike-share dock availability.

  • hard
  • Two Sigma
  • Machine Learning
  • Data Scientist

Forecast bikes available at a station

Company: Two Sigma

Role: Data Scientist

Category: Machine Learning

Difficulty: hard

Interview Round: Technical Screen

## Data Analysis / Forecasting Prompt You are given historical Citi Bike (bike-share) trip and station status data. Each station has a fixed dock capacity. **Task:** Build an approach to **predict the number of bikes available at a specific station** at a future time (e.g., 15/30/60 minutes ahead). ### What to cover - How you would define the prediction target (label) precisely. - What features you would use (time-based, station-based, demand/supply signals, external data, etc.). - How you would split data for training/validation/testing given time dependence. - What baseline(s) you would start with and how you would evaluate the model. - How you would handle practical issues such as capacity limits, missing data, and unusual events.

Quick Answer: This question evaluates proficiency in time-series forecasting, feature engineering, handling temporal data splits, and incorporating operational constraints for predicting bike-share dock availability.

Related Interview Questions

  • Analyze Temperatures and Update Regression - Two Sigma (medium)
  • How would you forecast bike demand? - Two Sigma (hard)
  • Predict Bike Dock Demand - Two Sigma (hard)
  • Predict bike demand and avoid overfitting - Two Sigma (hard)
  • How detect duplicate card records? - Two Sigma (medium)
Two Sigma logo
Two Sigma
Jan 6, 2026, 12:00 AM
Data Scientist
Technical Screen
Machine Learning
2
0
Loading...

Data Analysis / Forecasting Prompt

You are given historical Citi Bike (bike-share) trip and station status data. Each station has a fixed dock capacity.

Task: Build an approach to predict the number of bikes available at a specific station at a future time (e.g., 15/30/60 minutes ahead).

What to cover

  • How you would define the prediction target (label) precisely.
  • What features you would use (time-based, station-based, demand/supply signals, external data, etc.).
  • How you would split data for training/validation/testing given time dependence.
  • What baseline(s) you would start with and how you would evaluate the model.
  • How you would handle practical issues such as capacity limits, missing data, and unusual events.

Solution

Show

Comments (0)

Sign in to leave a comment

Loading comments...

Browse More Questions

More Machine Learning•More Two Sigma•More Data Scientist•Two Sigma Data Scientist•Two Sigma Machine Learning•Data Scientist Machine Learning
PracHub

Master your tech interviews with 7,500+ real questions from top companies.

Product

  • Questions
  • Learning Tracks
  • Interview Guides
  • Resources
  • Premium
  • For Universities
  • Student Access

Browse

  • By Company
  • By Role
  • By Category
  • Topic Hubs
  • SQL Questions
  • Compare Platforms
  • Discord Community

Support

  • support@prachub.com
  • (916) 541-4762

Legal

  • Privacy Policy
  • Terms of Service
  • About Us

© 2026 PracHub. All rights reserved.