About me
Blog
Analytics

Cristian Perez Brokate

I use data science tools to process multiple source of data and build robust data process pipelines. I build web applications with modern frontend javascript frameworks. I'm also in charge of IT management for large compute servers. In my free time I build IoT applications.

What I do:

Data modelling

  • Regression and classification algorithms for finance and insurance French companies.
  • Deep learning algorithms for object detection
  • LSTM models for time series prediction in robotics application

Data engineering

  • Collect data from different kinds of sources: social networks, APIs and websites.
  • Save large amounts of data coming at high intensity using a differed processing pipeline.
  • SQL and NoSQL database management (mongodb, neo4j, oracle, MySQL, SQLite)

System management

  • Application deployment using docker container system
  • Cloud computing management using
  • System performance monitoring using telegraf, influxdb and grafana.

Visualization

  • Interactive dashboard development using javascript libraries such as d3.js, dc.js
  • Web applications using modern frameworks.
More about me

Last articles

Magic wand gesture recognition us..

This article aims to demystify the implementation of machine learning algorithms into microcontrollers. It uses runs a TensorflowLite mode..

๐Ÿ“…  Jun 4, 2021 data science, programming
Hydroponic Agriculture Learning w..

New methodologies of horticulture based-on high-end technology are urgently required to transform the way in which the world is fed. In t..

๐Ÿ“…  Dec 7, 2020 programming
TheFifthDriver: Machine learning ..

FPGA implementation of a highly efficient real-time machine learning driving assistance application using a camera circuit.

๐Ÿ“…  Dec 7, 2020 programming
Writing notes with Vimwiki and Hu..

Vim is a simple and ubiquitous text editor that I use daily. In this article I show how to use Vim to take and publish diary notes using V..

๐Ÿ“…  Jul 1, 2020 programming
Reverse proxy management with Tra..

Traefik is a modern and dynamic reverse proxy with a native support with docker containers. This article compares Traefik with existing so..

๐Ÿ“…  Jun 7, 2020 system management
Muse: Mopidy web client with Snap..

Services like Mopidy and Snapcast are ideal to create a multiroom streaming audio player using devices like the RaspberryPi or android tel..

๐Ÿ“…  Apr 26, 2020 programming
Search posts functionality in AMP..

Good web reference is related with how fast your web page loads. Google works in a web format called AMP, which makes loading pages faster..

๐Ÿ“…  Apr 6, 2020 programming
Arduino MKR FOX 1200 for IoT and ..

Microcontrollers are adapted for simple task which don't need a lot of battery power. This post shows how to put in actions the Arduino MK..

๐Ÿ“…  Mar 8, 2020 programming
Image processing using serverless..

We created a simple project to show how to deploy a lambda function that returns the shape and size in bytes of an image by passing binary..

๐Ÿ“…  Feb 24, 2020 programming

More at my blog

Featured projects

API Personnes dรฉcรฉdรฉes

API for people who died in France from 1970. It has a bulk mode, to process csv files with multiple identities, which can be used to clean..

  typescript   express   swagger
IA Flash

This artificial intelligence application detects a vehicle, its make and model in a photo and a video. It's currently used by French Gover..

  python   opencv   pytorch   sphinx
Object detection application

This application implements state of the art algorithms for object detection with a frontend application to interact with results. It can ..

  python   opencv   angular   raspberrypi   chartjs
Muse

A mopidy web interface with snapcast support. The perfect excuse to use state of the art web frameworks and CI/CD workflows.

  svelte   sapper   bulma   github-actions   mopidy
Grafana system monitoring

I adapted multiple monitoring components to deploy easily a monitoring system using tools like Telegraf, Collectd, Influxdb and Grafana.

  grafana   docker   influxdb   telegraf   aws