In my career I have done a great number of technical tests. Some I merely did as exercises for myself knowing that I was probably not going to get the job, and hoped to merely expand my skillset. I've obviously grown as a developer over the years, but I'm pleased with most of the tests I have done.
However, at this point I hope that anyone who wishes to hire me will examine this first, before compelling me to add yet another example of: Django, Flask, or a simple algorithm.
This was a demonstration of how to build an ideal SDK for an API. This focused on lowering the barrier to entry using GUIs embedded in Jupyter. And, I demonstrated how you can use MyPy with metaprogramming. I went quite in-depth trying to achieve my desired result, and subsequently patched the IPython kernel to fix a major bug I found.
This was a pretty simple CRON parser. But it demonstrated project structure and use of PyTest.
Demonstrates the use of Cython and C++ to do a simple job.
Demonstration of building an RabbitMQ RPC based microservice on top of Nameko. Resises and rotates images in different processes with S3 as shared memory. Ran in docker-compose.
Demonstration of how I would set up a Django application for API usage. I find that many Django developers frequently add too many batteries because they don't know what is integral or not. For instance, you probably do not need CSRF middleware if you only have an API. This was a highly stripped down Django project, reducing cruft.
Demonstration of a Flask application that tells you when to sleep. Not much to say. Used a Dockerfile, SQLAlchemy and Marshmallow.
Probably my worst technical test as I missed the point. I used this time to learn about Django Channels and Websockets.
Little algorithmic test that I solved with a graph search algorithm.