Bleed AI Intern Test - Assignment PDF

This document describes an internship opportunity at Bleed AI, which consists of two startups - Bleed AI (Edtech) and Bleed AI Solutions. As an intern, the applicant would create computer vision applications that Bleed AI (Edtech) could use to create tutorials. Some example application ideas include intruder detection, facial enhancement, and color detection. By completing this internship, the applicant would gain experience working directly with the CEO and potentially get hired as an engineer at Bleed AI Solutions. To apply, the applicant must complete and submit one of two coding assignments on Google Colab involving face blurring in videos or color detection and analysis in images.

Uploaded by

saeed anwar

We take content rights seriously. If you suspect this is your content, claim it here.

Available Formats

Download as PDF, TXT or read online on Scribd

0% found this document useful (0 votes)

159 views7 pages

Bleed AI Intern Test - Assignment PDF

Uploaded by

saeed anwar

We take content rights seriously. If you suspect this is your content, claim it here.

Available Formats

Download as PDF, TXT or read online on Scribd

You are on page 1/ 7

Bleed AI Intern Test / Assignment

Who are We?

If you don’t already know, we at Bleed AI consist of two startups.

1: Bleed AI (Edtech)
2: Bleed AI Solutions.

At Bleed AI (Edtech), we’re responsible for creating high-tech tutorials in a form of blogpost and
youtube videos. These are not your everyday medium articles but are well-researched and
high-quality tutorials that are composed of unique applications and theoretical explanations and
exciting computer vision use cases.

In fact, you can check out our published blog posts here: http://bleedai.com/

You can also check out our Youtube channel with videos here:
https://www.youtube.com/c/BleedAI/videos

At Bleed AI Solutions we are responsible for building real-world applications for actual clients, it’s a
software services business focused in computer vision services.

Note: In many cases, clients of Bleed AI Solutions are interested in things we create at Bleed AI so
the applications/tutorials that we publish are of utmost importance.

What will this internship be About?

Whenever we create a tutorial at Bleed AI (Edtech), we need to create an exciting computer vision
application or a project first, this is where you will come in. You will be tasked with creating a
unique CV application that will require, research and coding. Once you create that, we’ll take it from
there in and format it and add explanations, etc.

For this particular internship, there are many interesting applications that we want you to develop
for Bleed AI (Edtech)
Some of these are
● An intrusion Detection system, to detect intruders coming in a restricted area, IP
camera stream + GUI
● Facial enhancement application.
● Creating a dominant Color Detection System + Creating a color distribution chart for
the image too.
● Pose language from face and body gestures using project Shuwa a gesture toolkit.
● Create a chrome extension that applies computer vision on images OR youtube
videos
● And More…

Why do the Internship For Us:

While working with us, you’ll be directly managed by Taha Anwar (CEO of Bleed AI) himself so you’ll
be learning a lot, furthermore, each application will pose a unique set of challenges which will
further hone your skills and if you get stuck we’ll guide you accordingly.

In the past we have successfully run a similar model, interns who did well for Bleed AI (Edtech)
were even hired as computer vision engineers to work for clients at Bleed AI Solutions. We’re
planning to do the same for this round of internships.

How do I get Selected:

Alright here’s what you need to do now.

You need to pass the screening test or assignment, once you complete it and submit it, we’ll then
evaluate the assignment, and based on your work, we’ll shortlist some candidates for the first few
weeks of paid probation.

After the paid probation period (in which you’ll also learn a lot) we will further evaluate you and
make the final selection. Please note, the internship will actually start with the probation period and
you’ll be working on the applications described above. The paid probation period will serve to tell
us if you can actually do the work.
Note: You only need to do one of the following assignments, you can do
either Assignment 1 OR Assignment 2. Assignment 2 is harder and so it
will have a higher score.

Assignment: 1 (Normal)
Create an application that takes a youtube video URL and blurs out all instances of Imran Khan’s
face present in the video.

This is a high-level workflow for the application:

Optional Helpful Resources:

Note: You can use these resources to complete the application, Make sure to cite at the end of the
notebook the resources you used. Also try to not blatantly copy-paste chunks of code from
someone’s blog post, since we can’t use that in our articles because of plagiarism. Using github
repos is fine.
● For face blurring:
https://pyimagesearch.com/2020/04/06/blur-and-anonymize-faces-with-opencv-and-pyt
hon/
● For face recognition: https://google.github.io/mediapipe/solutions/face_detection.html
● For reading youtube videos: https://github.com/mps-youtube/pafy
● Face detection: https://google.github.io/mediapipe/solutions/face_detection.html

Scoring Criteria:
● Your code should work and be error-free.
● You should Effectively solve the problem in most of the videos, your solution should be
robust and accurate.
● Your code should be clean, commented, and modular, create functions instead of writing
monolithic code.
● Adding descriptions in your code to explain how it works is even better.

Hint: You can download one of our notebooks from any of the blogpost we have published, this will
give you a good idea about how we structure our code. For e.g check the code for this blogpost.

How to Submit Your Assignment:

All your code should be in a google colab notebook, Before submitting your work, make sure to
restart the colab kernel and run it to make sure the code runs and it’s error-free, also at the end of
the notebook show that your code runs and blurs out the faces effectively and produces a final
video. You should include 2-3 sample outputs that show that your solution works accurately.

NOTE: Once you have finalized the colab notebook, you can submit it here using this form. The
deadline to complete the assignment is 13th November. Make sure your notebook is publicly
accessible otherwise we won’t be able to check your work.

Example:
Where Final_Solution() is the function that contains the main code of your solution

Assignment: 2 (Hard)
Create an application that takes an image and provides you a color distribution chart of that image
and also Optionally tells you the dominant color in the image.

Example:
For inspiration, this is exactly what Google Vision API does: Link Now I don’t expect you to create a
solution like google but let’s see what you can create.

Note: If your notebook requires data to be stored, then you can either use public google drive folders
or open source github repo and pull it inside the colab notebook, this is because the google colab kernel
is unable to store data in its server.
Scoring Criteria:
● Your code should work and be error-free.
● You should Effectively solve the problem in most of the videos, your solution should be
robust and accurate.
● Your code should be clean, commented, and modular, create functions instead of writing
monolithic code.
● Adding descriptions in your code to explain how it works is even better.
● Make sure you include references or links you used to develop the solution.

How to Submit Your Assignment:

All your code should be in a google colab notebook, Before submitting your work, make sure to
restart the colab kernel and run it to make sure the code runs and it’s error-free, also at the end, You
should include 2-3 sample outputs that show that your solution works accurately on different
images.

BEST OF LUCK