Home on Viny Brasil's blog

Day trading Simulator built with Raylib

Thu, 15 May 2025 00:00:00 +0000

(On mobile play the game in fullscreen)

Optimizing Python with Zig for numerical calculations

Mon, 28 Apr 2025 00:00:00 +0000

Introduction

As a interpreted language, Python is known for being slower than compiled languages when running almost all processes including numerical calculations. It’s a trade-off actually, since writing compiled languages is normally more complicated (having to deal with pointers, for example) than Python.

There’s a way of profiting from both of the good sides: using dynamically linked shared object libraries. Compiling a function used in the calculation from the compiled language code to a shared object and loading it in Python (where we’ll call the function with the parameters) is normally faster than just pure Python.

Caching API responses with Redis in Zig

Tue, 11 Mar 2025 00:00:00 +0000

Introduction

When fetching a database, our server must perform a READ operation. Every transation has a cost of time and money (computational cost in the cloud). One way of reducing this cost it’s to cache the last record fetched so we don’t have to perform another READ operation. This can help when the same record is requested multiple times.

To do this caching, we can use Redis, a advanced key-value store that works as a in memory database. It is really fast and combines with our choice for the backend: zap, a blazingly fast webframework written in Zig.

Running sklearn models in Zig

Thu, 06 Feb 2025 00:00:00 +0000

Introduction

Scikit-learn models became the industry standard for creating machine learning models. Although the library offers many conveniences, one becomes tied to Python APIs for serving such models. These frameworks (such as FastAPI, for example) have various scalability issues, as Anton demonstrates in this video. To overcome this limitation, one solution is to use Python’s C library to run the models, while the rest of the API can be built using another framework.

Building a real-time Websocket webapp with Kafka and Javascript

Mon, 06 Jan 2025 00:00:00 +0000

Introduction

Suppose we have a API that generates a lot of information whom we must store in a database and show them to clients. This information updates every few seconds and this sort of system is call a real-time app (in reality, it takes some seconds, so it’s a near real time app). Some applications that can use this sort of design are:

chatbots: the API is a LLM that receives an input and send it’s response to the client;

Convert decimal odds to probabilities with Zig

Mon, 30 Dec 2024 00:00:00 +0000

A Zig program for conversion and payout calculation

Introduction

As sports betting expands across the world, more and more players don’t understand tha math behind it. So, the purpose of this article it’s to clarify how to convert from decimal odds to the probability behind them. The calculations are implemented in Zig, a system programming language perfect to create CLI tools because it’s fast and simple.

This article it’s divided into two sections: first we’ll talk about the math and them about the implementation. If you only need the code you can go directly to my GitHub.

Trigger a AWS Lambda with a SQS message with Python

Sat, 03 Dec 2022 00:00:00 +0000

Introduction

In the context of microservices, the communication of them must be fast and realiabe. This communication can be through messages and a tool for this communication can be SQS (Simple Queue Service), a messaging service by AWS.

This project will be a example of a system that can receive a SQS message from a client, process it, and give the information to another client. As a example of this process, we’ll create a microservice in a AWS Lambda that will act when receive a SQS message. It’ll write the message in a .txt file and return all the messages when receives a HTTP call to a certain endpoint. We’ll be using AWS Lambda because it is a serverless framework, in which the idea is you don’t have to worry about servers.

Deploying a classification model on AWS Lambda with Docker and FastAPI

Wed, 31 Aug 2022 00:00:00 +0000

Does the user inserted the correct picture when the app requested? That’s a pretty common problem companies faces when creating a onboarding process of a app or a web service. For example, during the onboarding process of a bank the app can ask for a selfie and the user insert a picture of it’s ID. Since in those kind of processes there’s a huge amount of requests at the same time, a manual validation process can’t be used and a machine learning model can be a good alternative. TensorFlow and FastAPI can be used to create that solution.

An asynchronous Consumer and Producer API for Kafka with FastAPI in Python

Mon, 27 Jun 2022 00:00:00 +0000

Writing asynchronous code it’s worth the effort when you’re dealing with a high load or multiple microservices that can take some time to answer your calls. The purpose of this article is to create an simple asynchronous API that works the same time as a Kafka’s producer and consumer. The full project it’s on my Github.

Setting up Kafka

In this tutorial we’ll be using Docker to set up Kafka following the Shuyi Yang’s tutorial[1]. The docker-compose.yml file used will be:

About me

Mon, 01 Jan 0001 00:00:00 +0000

Work

I’m a Machine Learning Engineer working for over four years in the industry using machine learning in scalable systems to add value and solve business problems.

I use Python and AWS/GCP as my main work tools.

Research

Reinfocement learning: In my Master’s thesis I studied reinfocement learning, where I developed a new algorithm called QSVR. The algorithm uses a Support Vector Regression algorithm call Online-SVR to estimate the Q-function of a environment. The results show it can perform better than the Deep Q-Network and Q-learning in a environment with small state space and a discrete action space. The code of the project is in this repository.

My Youtube videos

Mon, 01 Jan 0001 00:00:00 +0000

Home on Viny Brasil's blog

Day trading Simulator built with Raylib

Optimizing Python with Zig for numerical calculations

Introduction

Caching API responses with Redis in Zig

Introduction

Running sklearn models in Zig

Introduction

Building a real-time Websocket webapp with Kafka and Javascript

Introduction

Convert decimal odds to probabilities with Zig

Introduction

Trigger a AWS Lambda with a SQS message with Python

Introduction

Deploying a classification model on AWS Lambda with Docker and FastAPI

An asynchronous Consumer and Producer API for Kafka with FastAPI in Python

Setting up Kafka

About me

Work

Research

My Youtube videos

Why you will lose money with sports betting (mathematically) [video in portuguese]