Summary
Overview
Work History
Education
Skills
Interests
Timeline
Generic
FELIPE ARAYA

FELIPE ARAYA

Senior Data Scientist
London

Summary

Personal Statement

Data Scientist with 5 years + of experience delivering data driven insights in machine learning and algorithm-based solutions within the mortgage, property and consultancy industry (fintech, proptech, and consultancy). Capable of creating replicable machine learning algorithms from scratch using only mathematical theory as well as diagnostics without depending on a particular library. Characterised for having a driven curious nature, constantly seeking the best solution for a particular problem. Especialised in time-to-event models such as Churn models, survival models, and next event forecasting models.

Key Skills

  • Extensive knowledge of mathematical theory behind all essential machine learning algorithms such as Multivariate Regression (Linear and multiple linear regressions), Logistic Regression, Feed Forward Artificial Neural Networks (for both classification and regression, Naives Bayes Classifiers, Support Vector Machines, Decision Trees, Random Forest and K-Means Clustering.
  • Extensive professional experience with Python, Numpy, Scipy, Pandas, Matploylib, Seaborn, Plotly.
  • Expert-level proficient using R with nearly all tidyverse packages, such as readr, dplyr, tidyr, broom, ggplot2, reshape2, forcats, rvest, and purr.
  • Experience working in production environment using solutions such as TensorFlow, PyTorch, Spark, Dask, Prefect, MLflow, Seldon, AWS, Azure, Snowflake, and Dataiku.

Overview

6
6

Years of professional experience

2
2
Languages
4
4

Number of industries with relevant experience

3
3

Years as a Senior Data Scientist

Work History

Senior Data Scientist

Outra Limited
London
02.2022 - Current
  • Outra is a data-driven property insight company that especialises in provinding clients with targeted data at a household level to focus their resources and delivery of their services.
  • Utilized advanced querying, visualization and analytics tools to such as KeplerGI, Seaborn and Dataiku to create maps and diagrams for non-technical users to visualise a certain problem.
  • Discovered new problems and solutions for our exisiting data and model assumptions that were impacting the performance of our models.
  • Promoted and automated QA prodecures to monitor the healthiness and stability of our data.
  • Worked on improving existing next event forecasting models by calibrating the models, incorporating a new scoring logic and increase the performance by 5x.
  • Created and implemented new next event forecasting models using survival analysis
  • Identified, measured and recommended improvement strategies for metrics across various models.
  • Presented findings orally and in writing with advanced mathematical models.
  • Worked on Dataiku crating roadmaps and models flows.
  • Worked on the migration from Dataiku to a new custom Intelligence fabric that using Open MLOps

Data Scientist

Belmont Green
08.2019 - 02.2022
  • Belmont Green is an specialist mortgage lending company who ambitions to become a bank in the near future
  • Building machine learning algorithms and statistical models for time series data, focusing on retention, lifetime value, expected loss models.
  • Taking ownsership of projects from start to finish to ensure proofs of concept were properly implemented and deployed in production.
  • Performed exploratory data analysis and generate visualisations using Python, Pandas, Matplotlib, Seaborn, NumPy, Scipy, and other tools to tell stories with data and provide valuable insights to project owners and business stakeholders.
  • Implement machine learning algorithms for a variety of tasks such as Cashflow models, Early redemption models, Default models, Pre-payment models and conversion models, using Python and R.
  • Using data for analysis from a variety of sources using SQL, Azure DevOps and SQLAlchemy.
  • Build incremental models to measure the effect of retention, customer life time value and expected loss for the business.
  • Perform clustering and segmentation algorithms to gain insights into usage and appeal of various products and features for marketing and other purposes.
  • Designed and implemented database with SQLAlchemy framework for encryption/ migration and queries.
  • Responsible for creating a conversion model from start to finish using surival analysis techniques.

Co-founder and Operations Director

Andinas
commerce
03.2016 - 07.2019
  • Retailer that brings back the unique native South-American culture through hand-made shoes from Peru.
  • As a Co-founder and CEO I was initially responsible for every aspect of this business, including marketing, finances, operations, supply chain which gave me a wider business perspective.
  • The company was hosted through Prestashop and used Hong Kong solution companies to handle imports from Peru and distribution to Europe.
  • Worked using tools such as Google Analytics to track business performance and KPI’s.
  • Involved in the marketing process using Facebook ads and Google ads.
  • Directing the sales through the online portal as well as talking to potential retail customers and individuals
  • Later, I directed data-related solutions using machine learning and automation tools such a RPA and Selenium.
  • Build and diagnostic Logistic regression and Neural Network models for forecasting key performance indicators using Python
  • Matplotlib, Seaborn, NumPy, Scipy, and other tools to tell stories with data for non-technical stakeholders.
  • Implement predictive models for e-commerce data aimed to predict customer bahaviours and improve buying rate.
  • Perform clustering and segmentation algorithms to gain insights into usage and appeal of various products and features for marketing and other purposes.

Data Scientist

Botster AI
01.2017 - 01.2019
  • Performing data exploration and analysis, as well as building machine learning algorithms and statistical models to drive data and evidence-based insights for designing new features for several Start-up and Medium size companies in the UK and USA
  • Build and diagnostic Neural Networks models for forecasting key performance indicators using Python and TensorFlow.
  • Provide insights from forecasts via live interactive dashboards created with plotly.
  • Take ownsership of projects from start to finish to ensure proofs of concept are/were properly implemented and deployed in production
  • Perform exploratory data analysis and generate visualisations using Python, Pandas, Matplotlib, Seaborn, NumPy, Scipy, and other tools to tell stories with data and provide valuable insights to project owners and business stakeholders
  • Implement machine learning algorithms for a variety of tasks on multiple teams using Python and/or R.
  • Access data for analysis from a variety of sources using SQL
  • Build incremental models to measure the effect of feature improvements/changes.
  • Perform clustering and segmentation algorithms to gain insights into usage and appeal of various products and features for marketing and other purposes.
  • Perform ethical web scraping using Python with scraPy, RoboBrowser, and BeautifulSoup to obtain data for various analyses.
  • Perform exploratory analysis of unstructured text data to find trends in user reviews and identify specific features for improvement.

Education

BSc - Business Management, Mathematics

Kingston University
09.2013 - 07.2016

BSc Civil Engineering - Mathematics

Adolfo Ibanez University
03.2011 - 07.2013

Skills

    Python

undefined

Interests

Artificial Intelligence

Anime

Basketball

Chess

Side Hustles

Timeline

Senior Data Scientist

Outra Limited
02.2022 - Current

Data Scientist

Belmont Green
08.2019 - 02.2022

Data Scientist

Botster AI
01.2017 - 01.2019

Co-founder and Operations Director

Andinas
03.2016 - 07.2019

BSc - Business Management, Mathematics

Kingston University
09.2013 - 07.2016

BSc Civil Engineering - Mathematics

Adolfo Ibanez University
03.2011 - 07.2013
FELIPE ARAYASenior Data Scientist