Constantin is available for hire

Constantin Catalin Craita

Verified Expert in Engineering

Data Scientist and Developer

Bucharest, Romania

Toptal member since February 1, 2022

Expertise

Linux Python TensorFlow Spark NLP SQL Deep Learning Hadoop

Bio

Constantin is a data scientist with three years of experience, having experience in natural language processing and banking data science. He is also proficient in Python, SQL, Linux, Spark, Hadoop, and Airflow. Relevant projects he was a part of were named entity recognition, income estimation, anti-money laundering, and personalized customer communication.

Portfolio

BCR

Python, SQL, Spark, Hadoop, Linux, Scikit-learn, MLflow, Pandas, Statistics...

Mediatel Data

Python, Deep Learning, TensorFlow, REST...

Experience

Python - 5 years
Linux - 5 years
SQL - 5 years
Generative Pre-trained Transformers (GPT) - 3 years
TensorFlow - 3 years
Natural Language Processing (NLP) - 3 years
Statistics - 3 years
Pandas - 2 years

Availability

Full-time

Preferred Environment

Linux, Visual Studio Code (VS Code), Jupyter Notebook, Jira, Confluence, Microsoft Teams, Google Colaboratory (Colab), TensorFlow, Pandas

The most amazing...

...thing I've developed is a zero-shot transfer learning in named entity recognition (NER) task from French and German to English.

Work Experience

Expert Data Scientist

2020 - 2021

BCR

Developed a model which matches similar names using CNN-based neural networks.
Built an anti-money laundering model based on transactional and customer data.
Created personalized communication based on customer segmentation and clustering algorithm.
Developed an income estimation model as a regression on demographical and transactional data.
Worked closely with data warehouse team and customer communication team.

Technologies: Python, SQL, Spark, Hadoop, Linux, Scikit-learn, MLflow, Pandas, Statistics, Deep Learning, Classification, Text Classification

Machine Learning Engineer

2019 - 2020

Mediatel Data

Developed an intent classification model based on a speech to text from user calls as an NLP deep learning model with RNNs.
Created an email classification model with additional information from OCR parsed images and PDF parsing mining for customer data from attachments.
Dealt with business meetings with clients and various PoCs.
Migrated TensorFlow model to C++ and REST API to integrate with rest of legacy code.

Technologies: Python, Deep Learning, TensorFlow, REST, Generative Pre-trained Transformers (GPT), Natural Language Processing (NLP), Speech to Text, Optical Character Recognition (OCR), Email Parsing

Experience

Named Entity Recognition

I created a named entity recognition (NER) NLP model based on the BERT language model and deep learning. I also used multiple datasets in French, English, German, and Romanian from various fields such as press, medical, and legal. In addition, I implemented state-of-the-art novel techniques such as zero-shot transfer learning, classical transfer-learning, multi-lingual dynamic word embeddings, and multi-task learning.

Skills

Libraries/APIs

TensorFlow, Pandas, Dask, Scikit-learn, Natural Language Toolkit (NLTK)

Languages

Python, SQL

Platforms

Linux

Frameworks

Spark, Hadoop

Paradigms

REST

Other

Statistics, MLflow, Deep Learning, Classification, Text Classification, Natural Language Processing (NLP), Speech to Text, Optical Character Recognition (OCR), Email Parsing, 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.

Share your needs

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

Choose your talent

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

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