I am an A.I. post-graduated and an experienced system analyst ( working in financial sector, banks, and exchanges) by designing solutions, developing programs as Sr Cobol code, and leading a multi-skilled team as a tech lead.
Now I am pursuing a career change to the A.I. world, programming in python and eating books about statistics and mathematicss.
I am currently working on a private personal project building a python lib for helping EDA in Jupyter notebook environments.
I am also starting a youtube channel to share my discoveries in data science as well as a blog/medium to share knowledge called sigma squared.
You can reach me by mail at fabiojr@skiff.com or sigmasquared@skiff.com
(Please, remember to take a look in my capstone project)
| Project | Description | Status |
|---|---|---|
| Airbot (DRL Stock Trading bot platform) | Final Academic project of my postgraduate degree in artificial intelligence where we explored several state-of-the-art AI techniques such as Deep Reinforcement Learning and Retrieval-Augmented Generation (RAG) to produce a proof-of-concepan for an automated stock trading platform where the system itself takes buying, selling and portfolio management on your behalf. | [x] Finihsed, but some of the repositories are private due the security of credentials and our data |
| Kitai Hakai | A simple chat-bot for telegram, light and customizable | -[x] beta version available |
| COVID-19 Detection | Detects COVID-19 in lung X-rays using Convolutional Neural Networks. This is an improved version of parts of a college group project I participated in. I extracted the parts I worked on and reworked the code to optimize it and focus on detection via CNN. | -[x] Jupyter Notebook finished |
| Slotermeyer | Python Lib for speeding up EDA process. For Kagglers analyzing data through EDA and descriptive analysis in the Jupyter environment. | -[x] Alpha Private Version |
| SigmaSquared | A repository for keeping the source codes from my channel project to share knowledge about A.I. through videos and articles. | -[ ] In progress |
This is a preview of my capstone project back-end where I am testing different models through trading historical data.
- This is not the final interface, just for back-end testing.
- This one is made using Gradio but my team and I built a responsive web platform using reace.js
Please refer to the Airbot repository to find all the sources for the project and the paper-like final report
From a pandas dataframe you can, with a simple command, generate a report on your data organized into sessions and tabs. Summary, Numerical data, and Categorical data, proving information and graphs using descriptive analysis techniques. Theses and many more functions for univariate and multivariate analysis, time-series analysis, and much more.
This way, the user can have information about their data with just a few lines of code, organized in a cleaner way within the Jupyter notebook. Instead of having to scroll through pages and pages, the user uses tabs and sections that are organized by features.
Find more party parrots here, all the birds here are from there. =]