DEVLog - June

DEVLog - June

Hey there, how is it going?

Many friends has asked me why I have not published during the whole month. The truth is I just was not in the mood, but after a short break I am here, ready to tell you what I have done during this month.

Bezier

I haven't done any progress with Bezier yet. Offline-first implementation has been harder than I thought. There are errors everywhere, nothing is working and this has been frustrating. Hopefully, last week I had something working, so if everything goes well from now on, I will be releasing a beta probably this month.

Follow my progress on X, and you will be notified before anyone else!

SignatureAPI

SignatureAPI is doing well, the dozens of sign ups week after week have become subscriptions, and those are amazing news in order to provide a digital signature API for high volume, and a low costs, that competes with giants like Adobe or Docusign.

This month I had to solve a lot of problems with glyph fonts, because despite we take how characters are displayed as a trivial problem, the truth is it is a pain when you do have to work with rendering systems. The good news are that everything is working and pretty soon you will see news on the SignatureAPI blog!

Electronic Signature Blog and SignatureAPI News - SignatureAPI
News about Electronic Signatures and SignatureAPI

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.

Deploying AI models on Cloudflare

The first class of the month, I showed my students how to deploy AI models using the AI workers of Cloudflare. There we deployed whisper and a llama3 model.

image.png

I also show them how to deploy models on Azure ML Studio, where we deployed a Microsoft-Phi-3 model.

image.png

AI friend

Then, I talked with my students about how to generate voice from text and as example we used Coqui-TTS. The project of the week was creating a real-time speech-to-speech chatbot as the shown by OpenAI the week before.

0:00
/0:40

Landing a Rover on the Moon

The class plan followed with Reinforcement Learning, and I took the opportunity to introduce my students to the world of Gymnasium and Stable Baselines. The project of this week was training an AI to land a rover on the moon!

0:00
/0:13

Gymnasium Workshop

Last week of the month was an assessment of the skills of my students to determine the main elements of a reinforcement learning problem and see if they were capable of using them in order to resolve some Gymnasium environments, seeing a little bit of everything!

The Next Steps

This month was awesome, and I expect more from July. Since the semester is almost reaching its end, I will have more time for other projects and experiences that I plan to post here in my blog, so if you want to receive updates about them, don't hesitate and join the newsletter!


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.