James Lee

Anderson Henrique de Oliveira Conceição

Junior Data Scientist

Send me a message!

About Me

I'm an engineering student turned data scientist who loves to analyze and create models based on data! I love to think about new innovative solutions in order to fix day-by-day problems, trying to reach the best experience for the customer.

My Projects

Regador de plantas Open Source

Projeto de fim de curso desenvolvido para a disciplina de Microcontroladores na Universidade de Pernambuco.

O objetivo do regador é, através de um sensor de humidade, manter a planta saudável ligando e desligando uma bomba (controlador on-off). O usuário pode informar o setpoint tanto localmente quando remotamente, através de um site.

View on GitHub

projeto_final_IA Open Source

Projeto de conclusão da disciplina de Inteligência Artificial, realizado por Anderson Henrique estudante da UPE em conjunto com a Secretaria da Fazenda de Pernambuco.

View on GitHub

Projeto de Regressão Múltipla usando Redes Multilayer Perceptron Open Source

Projeto realizado como trabalho de conclusão da matéria de Redes Neurais, ministrada pelo Prof. Dr. Mêuser Valença. O trabalho foi realizado por mim e pela minha equipe.

View on GitHub

ey_data_challenge_2019 Open Source

This project refers to my efforts in to the Data Challenge competition from EY.

View on GitHub

udacity_courses Open Source

Udacity Course Notebooks

View on GitHub

last-fm-history-analysis Open Source

I will be wandering through my last.FM music history and hope to discover meaningful insights.

View on GitHub

More on my GitHub!

Work Experience

Data Scientist Remote Intern - Dotz Marketing S/A - Brazil (09/2017 - 06/2019)

  • I worked as a remote intern at Dotz Next department, also known as the innovation and analytics section. Developing projects on demand and proofs of concept, which requires the construction of analytical models in Python and in R.
  • Knowledge and experience using Google Cloud Platform products to design and launch the developed projects. As my job is remote, does requires a certain discipline and self- organization in order to maintain my presence in the department.
  • Modeling and construction of a Recommendation System, for retail products using state-of- art algorithms. This project began in the Summer Job, but it was developed, increased performance and escalated. To be used in production. This solution has decreased costs for the company and has increased the value of our service to our partners.
  • I have developed a model for Profile Clustering of Customers to offer better insights for our partners and more data to understand the customer churn.

Summer Job Intern - C.E.S.A.R. - Brazil (07/2017 - 08/2017)

  • I have participated of the Summer Job 2017.2 edition from C.E.S.A.R., where a group of 4 graduation students is selected to develop a proof of concept as the solution for a problem proposed by a Sponsor’s company.
  • The team has then 6 weeks to build the solution, with the support of tutors and the infrastructure of the company. We have then received lectures related to the subject of the challenges.
  • The company that has sponsored us, propose the challenge of modeling and build a Recommender System using the customer history of retail bought products (Big Data).
  • The organization of the team was made by the method SCRUM; the idealization of the problem was made using design thinking. And in the solution proposed, we have used the Microsoft Azure.
  • The experience served as the first step of my career in data science. At the end of the program, the sponsor’s company hired me as an intern.

Measurement Error Internship - IMG GmbH - Germany (04/2015 - 07/2015)

  • The budget of the measuring errors for motor vehicles electromagnetic compatibility measuring stations is to be determined by simple means based on measurement location error descriptions from the literature and from existing measurement errors.
  • A study was made to the exclusive measuring chain, to understand 'how', 'where from' and 'why' the error can propagate along the processes.
  • The results of the measurements were discussed by practical tests in the high-frequency laboratory of the enterprise.

My GitHub chart

My GitHub chart