DEVLog - May

DEVLog - May

It's unbelievable that we are in June, the time has passed so quickly and it's time to review what I did.

Bezier

If you have been following on my social media accounts, you might notice that I haven't posted any updates as I used to.

The reason of this is that most of the time of this month the app has been down due some changes I have been implementing in order to achieve an offline-first experience in the app. This has required a lot of backend and internal logic that broke the application and all the progress I have had this month is related to make the app work as it was working in the beginning of May.

SignatureAPI

SignatureAPI has starting growing the last few weeks, having dozens of new users week after weeks since the completion of the deployment of the Microsoft Power Automate connector.

This month we expect to release an internationalization feature that has been asked for many of our customers and enhance many other flows.

If you are looking for a platform that provides a digital signature API for high volume, and a low costs, check out the SignatureAPI page!

SignatureAPI
Electronic signature API for workflows, applications and platforms

AI and ML

As many of you already know, I'm a professor of Artificial Intelligence and Machine Learning and this month I implemented many projects to teach my students the capabilities of some of the models we talk about.

RemoveBG

The first project of the month was removing background of images using the Segment Anything Model. This way we could identify the masks of the image in real time and remove the background mask.

0:00
/0:22

Image Upscaler

Continuing with autoencoders, I presented my students how to use existing models in Hugging Face and Replicate, and took as example the problem of upscaling ans super resolution of images.

0:00
/0:03

Denoiser

Starting with audio processing I showed my students how to create an audio denoiser tool using a SepFormer working on the time domain.

0:00
/0:47

Trader

Starting Recurrent neural networks, with my students I created a trader bot to predict Bitcoin prices using the Yahoo Finance API and a Simple RNN, LSTM and GRU, to predict the prices and comparing what model performed better. Then we used it to predict the price that BTC will have by the middle of the next month (72.000 USD), let's see how it goes!

Translator

The last project of the month was implementing a translator application using our own transformer, this way my students could learn the inner structure of a transformer, and were able to create an app to translate from Spanish to English, perhaps with an A1 level.

0:00
/0:27

The Next Steps

This month was wonderful, and June promises a lot of awesome projects and fun. I have been pretty busy, so I haven't created blog post as frequently as I would like to, but in 2-3 weeks I will finally recover a few hours in my life and I will create new post about offline-first applications, AI and ML applications, and miscellaneous events that happen in my journey. See you until then!


If you enjoy the content, please don't hesitate to subscribe and leave a comment! I would love to connect with you and hear your thoughts on the topics I cover. Your feedback is greatly appreciated! Leave a tip.