Learn Data Skills
Beta
Filipe Luz

Filipe Luz

Senior Data Scientist

Universidade Cruzeiro do Sul | Brasília, DF

Technologies

My Portfolio Highlights

My New Course

Advanced Deep Learning with Keras

My New Course

Statistical Simulation in Python

My New Course

Case Studies in Statistical Thinking

I ask (and answer) critical business questions that have never been addressed before, uncovering opportunities, and driving efficiency and results through Data Science | AI | ML | Deep Learning | Big Data | BI

My Work

Take a look at my latest work.

project

Dr. Semmelweis and the Discovery of Handwashing

course

Case Study: School Budgeting with Machine Learning in Python

My Certifications

These are the industry credentials that I’ve earned.

Other Certificates

Cambridge Assessment English Council of Europe Level B1

Udacity Deep Learning Nano Degree Progam

SAP C_BOWI_42 Application Associate - SAP Business Object Web Intelligence 4.2

SAP C_SPDDA_165 - Certified Technology Associate - Data Consultant for PowerDesigner 16.5

Microsoft Azure Data Scientist Associate - DP100

DataCamp Course Completion

Take a look at all the courses I’ve completed on DataCamp.

My Work Experience

Where I've interned and worked during my career.

Youse Insurance | Oct 2021 - Jul 2023

Senior Data Scientist

• Implement a new Total Loss pricing definition model. Apply Machine Learning algorithms (LGBM) to improve the conversion of clients in new contracts. This project improves 3% of contract conversions having a Gini parameter greater than the GLM model. • Maintain 5 real time models running in production (AWS Sagemaker) and its reports. • Analyze data from Bureaus to find better metrics that could reduce the risk uncertainty for claims. We analyze 3 bureau datasets, finding 1 feature that could reduce the uncertainty. • Implement A/B test environment, making possible compare elasticity adjusts for products Pricing. • Reduce the impact of loss ratio occurred in insurance market, using data, suggesting price increase or reducing, acting with the Actuarial team. • Implement alternative to score clients in using AWS Endpoints at QA Environment, making possible analyze profiles for market competitiveness. • Improve ETL process using Scala or PySpark.
Show More

Cast group | Oct 2018 - Present

Data Scientist

● Led a project to predict network traffic consumption for 3.500 Bank of Brazil branches, by applying ML and AI to avoid service unavailability, thus safeguarding ≈ US$ 1M/hour in revenue per branch. branch. ● Implemented an efficient incident channel to assist internal and external clients of the Bank of Brazil regarding systems issues using the Pytorch Bert Model. This increased productivity and decreased manpower need from 6 to 2 professionals to manage issues. ● Decreased the internal system unavailability by 15%, by employing anomaly detection, Kafka, Python, CI/CD using Jenkins, Argo, and Kubernetes, among other tools, avoiding an estimated profit loss of ≈ US$ 140K/hour due to the unavailability of financial contracts. ● Developed a Spotfire dashboard to demonstrate forecast outcomes the number of incidents for each Bank of Brazil application allowing managers to decide about the workforce distribution for the next week’s operation, thus increasing productivity.

First Decision | Jan 2013 - Oct 2018

Senior BI Analyst and Junior Data Scientist

● Led, developed, and managed projects, with a focus on improving end-user and customer functionality, increasing revenue and efficiency while reducing risk across +20 different projects. ● Implemented projects for clients, from gathering business requirements to the development of BI tools and dashboards, using SAP solutions to support companies in the decision-making process. ● Led a project to automate and implement rules by ETL and ML regarding payment processes of 31 social programs, avoiding overpayments and generating savings of ≈ US$ 4M/month in reimbursement of incorrect payments. ● Developed a project to implement 61 KPIs to support forecasting regarding a government mandatory payment using Python, SAP, ML, avoiding delays in the release of investments and injecting US ≈10B into the Brazilian economy. ● Participated in a project to forecast the demand of reloading money supply in the ATM using Spark and Python, focusing on reducing logistic costs by sending the team to restock the money when needed instead of keeping a routine schedule with unnecessary visits. ● Developed a new database to store data regarding passport issuance, applying Power Designing, SAP, and Oracle, avoiding loss of information and unavailability to issue passports nationwide.

My Education

Take a look at my formal education

Master of Business Administration, Data Analysis for BI and Big Data in n/aUniversidade Cruzeiro do Sul | 2018
College Diploma, Technology and System Development in n/aUniCEUB - Centro Universitário de Brasília | 2012

About Me

Filipe Luz

Highly curious Data Scientist with a strong ability to capture, process, analyze, and communicate complex data. Brings 10+ years of experience employing data process as well BI, Artificial Intelligence and Machine Learning.

Powered by

  • Work
  • Certifications
  • Courses
  • Experience
  • Education
  • About Me
  • Create Your Data Portfolio for Free