Cristian Cardellino, Developer in Córdoba, Cordoba, Argentina
Cristian is available for hire
Hire Cristian

Cristian Cardellino

Verified Expert  in Engineering

Generative Pre-trained Transformers (GPT) Developer

Location
Córdoba, Cordoba, Argentina
Toptal Member Since
May 29, 2021

Cristian holds a PhD in Computer Sciences, specializing in deep learning for natural language processing. He is part of the natural language processing group in the Faculty of Math, Astronomy, Physics, and Computer Sciences at the Universidad Nacional de Córdoba (UNC), Argentina. In addition to teaching data science courses, Cristian is also a research scientist at Mercado Libre, the largest eCommerce platform in Latin America.

Portfolio

Tappedout.NET
Apache Airflow, Django, JavaScript, React, Flask, Machine Learning...
MercadoLibre
Python 3, Machine Learning, Deep Learning, Natural Language Processing (NLP)...
Santex Group
Artificial Intelligence (AI), Natural Language Processing (NLP), Data Science...

Experience

Availability

Part-time

Preferred Environment

Linux, Python, PyTorch, Natural Language Processing (NLP), Artificial Intelligence (AI)

The most amazing...

...thing I've developed was a recommender system for a trading card game that had enormous success on the website, improving 80% over the original site's solution.

Work Experience

Freelance Machine Learning and Data Engineer | Full-stack Developer

2016 - PRESENT
Tappedout.NET
  • Developed a recommendation system for a trading card game.
  • Set up an Airflow environment to automate the site tasks, especially data-related tasks.
  • Helped move the site infrastructure to Docker and Docker Compose.
  • Developed a machine learning system for bots to automatically pick strategic cards based on one of the game's mechanics.
  • Architected a card embedding visualization platform with JavaScript and CSS.
  • Set up many machine learning models for different tasks as microservices via Flask.
  • Designed a visual application with React for the platform's user to build their decks.
  • Worked on a data model to detect spam bots from log files. Started by analyzing the data of many different logs and worked using clustering techniques to differentiate bots from humans.
Technologies: Apache Airflow, Django, JavaScript, React, Flask, Machine Learning, Recommendation Systems, Spark, Docker, CSS, Keras, TensorFlow, Artificial Intelligence (AI), NumPy, SciPy, Scikit-learn, Seaborn, Pandas, Jupyter Notebook, Pytest, MySQL, Data Science, Machine Learning Operations (MLOps), Data Engineering, JSON, Docker Compose, Apache Spark, PySpark, REST, Full-stack, Web Development, REST APIs, XML, APIs, API Development, Deep Learning, PyTorch, BERT, Microservices

Senior Researcher

2020 - 2023
MercadoLibre
  • Helped develop team objectives and direction as the 1st employee in the research team. Aided the overall construction of the team for future members.
  • Researched product representations using both natural language processing as well as computer vision techniques. Found multimodal representations of items in the marketplace.
  • Researched cross-region domain adaptation as the marketplace is available in 18 different countries in Latin America.
  • Conducted research with scalability in mind, as Mercado Libre is the largest marketplace in Latin America.
Technologies: Python 3, Machine Learning, Deep Learning, Natural Language Processing (NLP), Research, Computer Vision, BERT, Custom BERT

Freelance AI Consultant

2018 - 2023
Santex Group
  • Worked as a manager in an agreement between Santex and the National University of Córdoba for developing artificial intelligence (AI) applications in the industry.
  • Acted as a freelance consultant for the company on applications involving artificial intelligence (AI), data science (DS), and natural language processing (NLP).
  • Oversaw and acted as a consultant for the technologies provided by OpenAI to help develop a chatbot for internal use.
  • Carried out technical interviews for possible candidates in technologies such as Python and Django and AI, DS, and NLP.
Technologies: Artificial Intelligence (AI), Natural Language Processing (NLP), Data Science, Machine Learning, Data Engineering, OpenAI GPT-3 API, Chatbots, Microservices

Data Engineer

2021 - 2022
Honest Paws, LLC
  • Helped the client identify problems in their codebase, fixing bugs and refactoring for better solutions, using Python, SQL, and Docker alongside the Google cloud services.
  • Developed many Google cloud functions needed for the client to manage their data pipeline to provide their business analysts with clear views of the data.
  • Assisted the client by teaching them best practices and explaining different use cases of their available technologies.
  • Created several cloud functions using Google Cloud that connected their data pipeline, in Google BigQuery, to a dashboard via Mixpanel.
  • Used a pipeline that applied the Google Cloud Natural Language Processing tool and wrote the results seamlessly for local storage and Google Cloud Storage.
Technologies: SQL, Python, Pandas, NumPy, Google BigQuery, Mixpanel, Python 3, Google Cloud, Google Cloud Storage, Docker, Google Cloud Functions, Google Cloud ML, Data Engineering, REST, REST APIs, Microservices

Adjunct Professor

2015 - 2022
Universidad Nacional de Córdoba
  • Taught students about SQL and MongoDB as a professor of database courses.
  • Led classes as a professor of Programming Paradigms, where I trained students on functional programming, object-oriented programming, and actor-based concurrency in Haskell and Scala.
  • Instructed Supervised Machine Learning, Deep Learning, and Recommendation Systems as part of the school of data science and machine learning specialization courses.
  • Researched many NLP topics, particularly text classification, named entity recognition (NER), sentiment analysis, and argumentation mining, among others.
Technologies: Databases, Scala, Akka, SQL, MongoDB, Machine Learning, Natural Language Processing (NLP), Generative Pre-trained Transformers (GPT), Haskell, Object-oriented Programming (OOP), Artificial Intelligence (AI), NumPy, SciPy, PyTorch, Deep Learning, Keras, Natural Language Toolkit (NLTK), SpaCy, Gensim, fastText, Word2Vec, MySQL, Text Mining, Entity Extraction, Named-entity Recognition (NER), Text Classification, Sentiment Analysis, Argument Mining, Technical Writing, XML, Chatbot Conversation Design, Microservices

Senior Full-stack Developer

2021 - 2021
Xofia
  • Designed a search engine for keywords related to the oil extraction industry. The keywords were extracted from HTML files and searched via PostgreSQL full-text search.
  • Implemented the search engine with an API via Python FastAPI, and the web interface using Django. Wrapped all the app with the help of Docker and Docker Compose.
  • Deployed the application as a droplet in Digital Ocean with the help of Docker in order to have a beta version for the client to showcase.
Technologies: Python, FastAPI, Django, PostgreSQL, Beautiful Soup, Docker, Docker Compose, DigitalOcean, Search Engines, Information Retrieval, Text Mining, REST, Full-stack, Web Development, REST APIs, Microservices

Intern

2014 - 2014
Inria
  • Developed a website to help with the automatic licensing of datasets.
  • Handled everything from design and front end to the back end.
  • Developed a natural language processing suite to give an RDF representation of a written license automatically.
Technologies: Scala, Play Framework, Natural Language Processing (NLP), Generative Pre-trained Transformers (GPT), JavaScript, jQuery, Flask, Python, Scikit-learn, SpaCy, Natural Language Toolkit (NLTK), Gensim, NumPy, SciPy, Machine Learning, Artificial Intelligence (AI), Technical Writing

Teaching Assistant

2011 - 2013
Universidad Nacional de Córdoba
  • Worked as a teaching assistant in three different courses: Programming Paradigms, Networks, and Software Engineering.
  • Oversaw and graded students' deliverables for the three courses, the Networks and the Software Engineering Courses.
  • Checked, tested, and provided feedback to the students regarding improvements to the software engineering course project completed in Django.
  • Reviewed the networks course deliverable, a complete peer-to-peer (P2P) system for data transfer written in Python.
  • Assisted the Programming Paradigms course professor in the practical part where we saw different programming paradigms such as declarative, functional, imperative, and object-oriented programming.
Technologies: Python, Django, Linux, Technical Writing, XML

Junior Full-stack Developer

2012 - 2012
Machinalis
  • Helped develop the ContinoIN platform in Django and Python.
  • Tracked and fixed many bugs using Jira as a reporting tool.
  • Wrote unit tests for Python and Django with the unit test framework.
  • Applied a design given to me by the website owners with HTML, CSS, and Bootstrap.
Technologies: Django, Python, jQuery, JavaScript, Bootstrap 3, CSS, Full-stack, Web Development

Recommendation System for MTG Site

https://www.tappedout.net
I was in charge of tappedout.net's deck recommendation system. I developed a microservice-style framework with Flask that runs a RESTful API that, given a set of cards, runs a hybrid recommendation system based on collaborative filtering and content-based filtering using scikit-learn and Surprise. I was in charge of the data extraction from the database, model training, model testing, API design and building (with security layer), and testing.

Spanish Billion Word Corpus and Embeddings

https://crscardellino.ar/SBWCE/
This resource consists of an unannotated corpus of the Spanish language of nearly 1.5 billion words, compiled from different corpora and resources from the web, and a set of word embeddings created from this corpus using the Word2Vec algorithm provided by the Gensim package. These embeddings were evaluated by translating to Spanish Word2Vec’s word relation test set.

Licentia

http://licentia.inria.fr/
As part of my internship at INRIA in France, I built the website Licentia, a suite of services to support users looking for a suitable license for their datasets. The application has four main components: find a license for the data given some restrictions, check if a license is compatible with the user's intentions, visualize and download and RDF license, and convert a written license into RDF representation.

Languages

Python, Scala, SQL, Python 3, JavaScript, XML, Haskell, CSS, C

Frameworks

Django, Flask, Spark, Scrapy, Apache Spark, Akka, Play Framework, Bootstrap 3

Libraries/APIs

Scikit-learn, PyTorch, Pandas, NumPy, React, Keras, TensorFlow, SpaCy, Flask-RESTful, SciPy, Natural Language Toolkit (NLTK), Beautiful Soup, PySpark, REST APIs, API Development, jQuery

Tools

Named-entity Recognition (NER), Apache Airflow, Gensim, Pytest, Seaborn, Docker Compose

Paradigms

Data Science, Microservices, Object-oriented Programming (OOP), ETL, REST

Platforms

Linux, Docker, Jupyter Notebook, Mixpanel, DigitalOcean

Storage

Databases, MySQL, MongoDB, PostgreSQL, JSON, Google Cloud, Google Cloud Storage

Other

Deep Learning, Natural Language Processing (NLP), Machine Learning, Recommendation Systems, Artificial Intelligence (AI), Word2Vec, Data Engineering, Machine Learning Operations (MLOps), Text Mining, Text Classification, Word Embedding, APIs, Generative Pre-trained Transformers (GPT), Algorithms, Data Structures, fastText, FastAPI, Entity Extraction, Sentiment Analysis, Technical Writing, Full-stack, Web Development, Web Scraping, Data Scraping, Natural Language Understanding (NLU), BERT, Custom BERT, Formal Methods, Languages, Google BigQuery, Google Cloud Functions, Search Engines, Information Retrieval, Argument Mining, Google Cloud ML, Chatbot Conversation Design, OpenAI GPT-3 API, Chatbots, Research, Computer Vision

2013 - 2018

PhD in Computer Science

Universidad Nacional de Córdoba - Córdoba, Argentina

2008 - 2013

Master's Degree in Computer Science

Universidad Nacional de Córdoba - Córdoba, Argentina