Leandro J M Machado
Verified Expert in Engineering
Machine Learning Operations (MLOps) Developer
Leandro is a data scientist and since 2013, he's been delivering data and machine learning solutions for clients in the fields of eCommerce, advertising, insurance, and financial services. His expertise ranges from early-stage prototyping to deploying large-scale machine learning services.
Portfolio
Experience
Availability
Preferred Environment
Amazon Web Services (AWS), GitHub, Docker, Spark, Scala, Python
The most amazing...
...thing I've helped to build was a digital bank from scratch for one of the largest banks in Brazil.
Work Experience
Lead Scientist
Vality
- Developed and maintained a credit-and-fraud analysis pipeline.
- Wrote risk, compliance, portfolio management, and pricing policies.
- Built and supported a proprietary credit score system based on bureau, purchase, and social network data.
- Mapped and integrated data providers.
- Led teams.
Data Scientist | Delivery Lead
Capco
- Developed and managed the delivery of analytics solutions for Bradesco's Next Bank; including project scoping, planning, reporting, and risk management.
- Led the development of a self-service help solution for Next Bank's mobile app. The model was based on TF-IDF information retrieval with feedback boosting using Thompson sampling.
- Managed the development of a spending limit recommendation solution for Next Bank's mobile app. The model was based on user-user collaborative filtering using account and credit card data with a linear regression model to handle a cold start.
- Led the development of a real-time offer recommendation solution for Next Bank's mobile app. The model had to track customer transactions in real-time to recommend personalized offers and perks based on customer preference and eligible partners.
Big Data Scientist
Chaordic
- Researched and developed recommendation algorithms for eCommerce.
- Developed a real-time product recommendation service for cross-store advertisement. The model used a customer's list of current interests to recommend similar products that are cheaper in other online stores. The solution also included a design of an auction system for offer placement bidding.
- Built a real-time bidding system for advertisements in order to automate and increase the ROI for the advertiser. The model considered a customer's navigation and purchase history to estimate the optimal bid price.
- Created an experimentation platform to streamline the evaluation and deployment of A/B tests. The platform helped reduce—by half—the time and costs of conducting online experiments.
Experience
Advice for Applying Machine Learning
https://medium.com/@leandromachado_11293/from-magic-to-method-advices-for-applying-machine-learning-fb363136e786Consumer Spending Limit Recommendation
The solution was built in Scala/Spark and leveraged terabytes of customer credit card and checking account transaction data. By using a combination of linear models and user-user collaborative filtering, the solution was robust enough for all classes of customers.
Real-time Credit Analysis Pipeline
The solution was implemented in Python and leveraged data from credit bureaus and other alternative data sources to efficiently evaluate customer creditworthiness. The fraud analysis step included facial recognition and matched the provided information with online sources of reputation such as social networks. With this combination of technologies, we could provide a fast-and-light onboarding process for the customer while achieving above market metrics of KS and Gini for default prediction.
Experimentation Platform
It consisted of a Scala/Spark job that took input from several sources—such as impressions, clicks, purchases, and so on—and processed this information to output a number of statistics and hypothesis tests. The methodology included an implementation of the Bag of Little Bootstraps T-test for a robust-and-computationally-efficient analysis. It reduced the time and cost of running experiments by half.
Recommendation System for Real-time Advertisements in eCommerce
The service was implemented in Scala and made use of variables such as product-content, product-CTR/Conv, and the customer's transaction history to recommend similar products. The solution was built in Scala and consumed data from databases such as Elasticsearch, Cassandra, and Redis. It also served over 30 advertisers and was deployed on one of the top three eCommerces in Brazil.
Skills
Languages
Python, Scala
Frameworks
Spark, Flask, Hadoop
Libraries/APIs
Scikit-learn, Spark Streaming, NumPy, TensorFlow, spray
Platforms
Jupyter Notebook, Amazon Web Services (AWS), Docker, Apache Kafka, Amazon EC2, VMware Tanzu Application Service (TAS) (Pivotal Cloud Foundry (PCF))
Storage
Elasticsearch, Amazon S3 (AWS S3), Apache Hive, Cassandra, Redis
Other
Recommendation Systems, Advertising, Data Analytics, Machine Learning, Machine Learning Operations (MLOps), Credit Risk, eCommerce, Gemfire, Cloud Foundry, Natural Language Processing (NLP), Social Networks, Credit Modeling, Fraud Prevention, Deep Learning, GPT, Generative Pre-trained Transformers (GPT)
Tools
Solr, GitHub
Education
Bachelor of Science Degree in Computer Science
Universidade Federal de Santa Catarina - Florianópolis, Brazil
Completed Credits in Computer Science
University of Virginia - Charlottesville, VA, USA
How to Work with Toptal
Toptal matches you directly with global industry experts from our network in hours—not weeks or months.
Share your needs
Choose your talent
Start your risk-free talent trial
Top talent is in high demand.
Start hiring