David Sainz, Developer in Dubai, United Arab Emirates
David is available for hire
Hire David

David Sainz

Verified Expert  in Engineering

Data Scientist and Developer

Dubai, United Arab Emirates

Toptal member since July 25, 2022

Expertise
Machine LearningSQLNeural NetworkNumPyRecommendation SystemsPythonA/B TestingDeep LearningNLPR
Bio

David is an experienced data scientist and software and algorithm developer, passionate about new technologies. He started coding when he was eight and has never stopped evolving his tech skills. He has a solid background in .NET, Java, Python, R, and C++ and has proven expertise in machine learning and data analysis. Despite being a self-driven and autodidact professional, David believes the most significant achievements are made in collaborative environments.

Portfolio

Uber
Python, Machine Learning, Data Science
Citi
Python, Data Science, Machine Learning
Astral Vision
Signal Processing, Algorithms, Machine Learning, C#, Node.js

Experience

  • SQL - 15 years
  • Data Science - 7 years
  • Machine Learning - 7 years
  • Pandas - 5 years
  • NumPy - 5 years
  • Python - 5 years
  • Recommendation Systems - 4 years
  • Neural Networks - 4 years

Availability

Part-time

Preferred Environment

Windows, Slack, PyCharm, Jupyter Notebook, Git, Zoom

The most amazing...

...machine learning project I've recently developed is a state-of-the-art recommender system for Citi, Wall Street.

Work Experience

Senior Data Scientist

2021 - 2022
Uber
  • Created a personalization system for the mobile app front page.
  • Built machine learning models for credit card underpayments and fraud.
  • Extracted stakeholders' needs into data science requisites.
  • Performed anomaly detection processes to detect fraudulent behavior.
Technologies: Python, Machine Learning, Data Science

Senior Data Scientist

2018 - 2020
Citi
  • Created different fintech machine learning models for default payment prediction or mortgage prepayment prediction.
  • Built a machine learning model for a loan recommendation engine.
  • Performed data analytics for finance analysis using Spark and Python.
Technologies: Python, Data Science, Machine Learning

Senior Algorithm Developer

2016 - 2018
Astral Vision
  • Performed data analysis and pattern extraction of virtual reality ride data.
  • Developed virtual reality algorithms for spatial tracking with signal processing.
  • Participated in the development of a virtual reality engine in C#.
Technologies: Signal Processing, Algorithms, Machine Learning, C#, Node.js

Data Scientist and Algorithm Developer — Marie Curie Fellowship

2011 - 2016
Technion Israel Institute of Technology
  • Performed big data analysis and graph processing of social network analysis (SNA) using R.
  • Implemented statistical data analysis of WiFi traces and file transfers for usage patterns.
  • Developed complex synchronization protocols for data consistency.
  • Created an algorithm for mobile data backup using commonly encountered devices around the user via mist computing.
Technologies: Java, C#, Distributed Computing, Data Analytics, Data Science, Algorithms, R

Data Scientist

2007 - 2010
Telefónica
  • Created machine learning models for predicting customer churn.
  • Performed large-scale data analysis and graph processing using R and MapReduce.
  • Analyzed data for social network analysis (SNA) to create a social graph out of call and message records.
Technologies: R, Java, C#

Experience

Credit Card Underpayment Detection

A real-time machine learning model to detect when credit cards will incur underpayment by calculating the probability of a card not having enough funds for payment and redirecting it to credit card pre-authorization. I took care of the whole funnel from data acquisition to deployment, retraining, and metric follow-up.

Mobile Front Page Personalization

Personalized a mobile front page for widgets and ads according to users' analyzed preferences and using contextual bandits and a popularity-usage mechanism to fall back to. It is a trained machine learning model using past data of user sessions and their clicks.

Mortgage Prepayment Predictors

Used machine learning to predict the prepayment rate of mortgage-based securities. The model was used to evaluate the quality of those securities and their evolution over time and developed using a deep learning approach with long-short term memory (LSTM) networks.

Loan Recommender

Used neural networks to create a recommender system for loans to investors. The system gives out loans per user ranked by relevance. After training the model, embeddings of users and loans are extracted from the neural net for further statistical analysis of client preferences and loan similarities.

Education

2011 - 2016

PhD in Computer Science

Technion Israel Institute of Technology - Haifa, Israel

2006 - 2008

Master's Degree in Computer Science

University of the Basque Country - Bilbao, Spain

Certifications

AUGUST 2020 - PRESENT

Structuring Machine Learning Models

Coursera

AUGUST 2020 - PRESENT

Neural Networks

Coursera

AUGUST 2020 - PRESENT

NLP and Sequence Models

Coursera

AUGUST 2020 - PRESENT

Convolutional Neural Networks

Coursera

AUGUST 2020 - PRESENT

Deep Learning

Coursera

SEPTEMBER 2019 - PRESENT

TensorFlow

Udemy

Skills

Libraries/APIs

Pandas, NumPy, Scikit-learn, Keras, Matplotlib, CatBoost, XGBoost, PySpark, TensorFlow, Node.js

Tools

Slack, PyCharm, Git, Skype, Zoom, Spark SQL, Jupyter, Apache Airflow, Redash, Seaborn, TensorBoard

Languages

SQL, Python, C#, Java, R

Paradigms

Distributed Computing, Agile, Scrum, Kanban

Frameworks

Presto, LightGBM

Platforms

Windows, Jupyter Notebook, Amazon Web Services (AWS)

Other

Machine Learning, Data Science, Research, Classification Algorithms, Random Forests, Regression, Long Short-term Memory (LSTM), Neural Networks, Dimensionality Reduction, Singular Value Decomposition, Recommendation Systems, Contextual Bandits, Reinforcement Learning, A/B Testing, Vowpal Wabbit, Deep Learning, Principal Component Analysis (PCA), Clustering, K-means Clustering, DBSCAN, Convolutional Neural Networks (CNNs), Natural Language Processing (NLP), Sequence Models, Deep Neural Networks (DNNs), IT Project Management, Signal Processing, Algorithms, Data Analytics, Clustering Algorithms, Generative Pre-trained Transformers (GPT)

Collaboration That Works

How to Work with Toptal

Toptal matches you directly with global industry experts from our network in hours—not weeks or months.

1

Share your needs

Discuss your requirements and refine your scope in a call with a Toptal domain expert.
2

Choose your talent

Get a short list of expertly matched talent within 24 hours to review, interview, and choose from.
3

Start your risk-free talent trial

Work with your chosen talent on a trial basis for up to two weeks. Pay only if you decide to hire them.

Top talent is in high demand.

Start hiring