Many people use AI to help them write documents particularly professional documents, but using standard online tools for writing is too cumbersome involving many copy/paste operations and tend to generate texts that are to generic and cannot be used directly before extensive manual editing. In order to help is this type of task, we have developed AIWrite, a very simple text generator using your choice of AI agent, that not only lets you properly and easily specify the structure of technical documents, but also makes it very easy to provide the AI information of your specific area of expertise, in the form of a collection of reference documents, so that the AI gent will write like a pro, not like a newbie!

We have just created an organization page on Hugging Face. You can find us at DeeplearnRnD. We are planning to release some of our models there, so stay tuned for updates! Feel free to follow us there as well!

We have just release a new version of Reggie D. Bot, v0.6.0. This release includes the following features: Improved user interface, with a new design and better user experience. Log of code snippets: Now every code snippet generated by Reggie D. Bot is saved in a log, so that you can easily access it later. The log entry is stored alongside the question that motivated it and an explanation. Many bugfixes and performance improvements.

After a few months of hard work, we are pleased to announce that we have made the first release of Reggie D. Bot. The demo is available at regdbot.deeplearn.ltd. It is still an alpha release, So at this point you can’t upload new datasets or register as an user. This release is for testing purposes only, and is linked with our project in the Backdrop build v5 program. Next steps before having a beta release, are to add the ability to upload new datasets, and to register as an user.

In our previous post we mentioned that we were joining the Backdrop Build program this year where we are proposing to turn our REG AI into a product. Even with our best efforts we felt that the Reggie D. Bot was not ready to be a product yet. So we rejoined Backdrop Build v5 to continue our work on Reggie D. Bot, while benefitting of building withing the backdrop community.

Our journey has just begun in Backdrop Build v4! Israel and I are super excited and we are looking forward to having a first release very soon. Today I have managed to prepared the demos database that we are going to use for the first release. It has 20 years worth of dengue incidence data for Reggie to chew on! Another exciting milestone is that we’ll have the first usable interface for our product out for testing by the end of the week.

Deeplearn has recently started a joint-venture with Publicase, a company that specializes in helping sciecntists write better scientific articles. The goal of this joint-venture is to develop an AI model that can act as an interactive writing assitant for scientists. The AI model will be able to help scientists write better scientific articles by providing suggestions on how to improve their writing. The AI model will be trained on a large dataset of scientific articles, selected by the the author, and will be able to provide suggestions on grammar, style, and content.

Deeplearn is joing the Backdrop Build program this year where we a proposing to turn our REG AI into a product. Joining efforts with us in this challenge is Israel Teixeira from NAN systems.

Deeplearn is joing the Backdrop Build program this year where we a proposing to turn our REG AI into a product. Joining efforts with us in this challenge is Israel Teixeira from NAN systems. Our challenge is to make our AI agent Reggie D. Bot into a product that can be used by the general public. We are excited to be part of this program and we are looking forward to the results of this challenge.

2023 is upon us! In this new Year, DeepLearn starts as a new external face to a team os experienced researchers with many years of experience of bringing AI to verticals typically ignored by big AI corporations. Along with the usual challenges related to simple survival, we are driven to find and develop new AI solutions than can solve problems instead of creating them (like the Fake news industry). In this road we hope to continue to meet inspiring people that can help us identify the worthy problems and the talented ones that will partner with us to build the solutions.