Where can you find qualified {role_name}?

What are the best practices for headhunting {role_name}?

How to shortlist candidates?

Once you have started to get applications from applicants, a thorough screening process and shortlisting of prospects will help you make the most of your time spent with the most qualified ones. 

Automated shortlisting tools :

Automated screening quickly filters out unqualified candidates, saving time for manual review. This allows the manual process to focus on the most promising candidates, ensuring the best ones are considered for further evaluation.

Screening questions to auto-shortlist based on predefined criteria

like qualifications, location, experience, and skills. Either use job board or use an ATS such as whitecarrot. Here are some questions for {role_name}

• How many years of experience do you have with computer vision algorithms?
  • Less than 1 year (Auto-reject)
  • 1-3 years
  • 3-5 years
  • 5+ years (Preferred)
• How many years of experience do you have working with Python and OpenCV?
  • No experience (Auto-reject)
  • Less than 1 year
  • 1-3 years
  • 3+ years
• Are you located within [specified location] or willing to work remotely?
  • Yes
  • No (Auto-reject if location-dependent)
• Do you have experience developing machine learning models for image classification or object detection?
  • Yes
  • No (Auto-reject)

Skill based question to auto shortlist candidate

Analyze the skill test data to automatically shortlist top-performing applicants. (recommended screening test time - 15 minutes). Here are some skill test questions for {role_name}

Computer Vision Algorithms

Testing knowledge of key computer vision concepts.

• Question: Which of the following is used for edge detection in images?
  • A) Gaussian Blur
  • B) Sobel Filter (Correct Answer)
  • C) Histogram Equalization
  • D) Image Resizing
• Question: What is the purpose of the Hough Transform in computer vision?
  • A) Object Detection
  • B) Line Detection (Correct Answer)
  • C) Image Segmentation
  • D) Feature Extraction
• Question: Which method is commonly used for feature extraction in image processing?
  • A) Convolutional Neural Networks (CNNs) (Correct Answer)
  • B) Principal Component Analysis (PCA)
  • C) Support Vector Machines (SVM)
  • D) Decision Trees

Machine Learning

Assessing the candidate's understanding of machine learning applications in computer vision.

• Question: Which machine learning model is most commonly used in computer vision tasks such as image classification?
  • A) Random Forest
  • B) K-Nearest Neighbors
  • C) Convolutional Neural Network (CNN) (Correct Answer)
  • D) Support Vector Machine
• Question: What is a key advantage of using transfer learning in computer vision?
  • A) Reduces the training time and data required (Correct Answer)
  • B) Increases the model’s accuracy by 100%
  • C) Automatically fine-tunes model parameters
  • D) Provides better visual outputs
• Question: In object detection, which method is used to generate bounding boxes around objects in an image?
  • A) YOLO (Correct Answer)
  • B) SIFT
  • C) HOG
  • D) RANSAC

Programming (Python, OpenCV)

Testing practical programming knowledge in Python and OpenCV.

• Question: Which OpenCV function is used to read an image from a file?
  • A) cv2.imread() (Correct Answer)
  • B) cv2.loadimage()
  • C) cv2.read()
  • D) cv2.showimage()
• Question: What does cv2.Canny() in OpenCV do?
  • A) Detect edges in an image (Correct Answer)
  • B) Resize an image
  • C) Convert an image to grayscale
  • D) Blur an image
• Question: Which Python library is commonly used alongside OpenCV for numerical operations?
  • A) Pandas
  • B) NumPy (Correct Answer)
  • C) Matplotlib
  • D) Scikit-learn

Note - Auto reject candidates if scores less than 70% in this section

One way video interview

Use tools like hirevue, whitecarrot.io to ask candidates pre-recorded questions about their experience and skills.

Use sample question given in scorecard.

Collect other information 

Collect data from shortlisted candidates, such as salary expectations and visa status.

Manual candidate profile shortlisting:

Thoroughly review the CVs of the top scoring candidates from the automated process

Look for evidence of the required skills, experience, and achievements

Review the candidate’s portfolio or GitHub repositories to see examples of their work.