Lucas Pavanelli

Lucas Pavanelli

About

Detail

AI Research Engineer | M.S. in Computer Science
Rio de Janeiro, State of Rio de Janeiro, Brazil
Contact Lucas regarding: 
work
Full-time jobs

Timeline

work
Job
school
Education
auto_stories
Publication

Résumé

Jobs verified_user 0% verified
  • J
    Part-Time Data Scientist
    Jobzi
    Jan 2020 - Jun 2021 (1 year 6 months)
    • Jobzi is a tech startup based in Rio de Janeiro that aims to provide smarter ways of matchingjobs and candidates using machine learning techniques. • Worked on a resume parsing tool applying machine learning algorithms (e.g., HMM, gradient boosting) to segment a CV into sections like Personal Data, Work Experience, and Education. (Java and Python)
  • J
    Data Scientist/Software Engineer Intern
    Jobzi
    May 2017 - Apr 2019 (2 years)
    • Implemented search algorithm using Solr. Improved search time by 50% and results quality by 30%.(Java and Python) • Used NLP techniques to recommend skills for resumes. (Python) • Worked onjob's salary identification using gradient boosting and metrics such as precision and recall to evaluate the proposed method. Improved recognition from 70% to 90%. (Java and Python)
  • Coursera
    Writing in the
    Coursera
  • a
    AI Research Engineer
    aiXplain
    • Built aiXplain's agentic framework based on LLMs calling external tools and made it available on SDK. • Led the development of a product called FineTune, allowing users to select their dataset and fine-tune different NLP models (LLM, MT, ASR) using suppliers such as HuggingFace, OpenAI, and Azure. • Built Flyte workflow that orchestrates LLM fine-tuning process and deployed on Kubernetes cluster for production-ready use. • Worked on ACL 60-60 initiative by automatically translating and evaluating over 10K ACL papers into 60 languages. • Developed ASR classification models to detect age and Arabic dialects. • Co-authored two patents and three scientific publications with the Science team.
Education verified_user 0% verified
  • S
    Sciences
    Jan 2021 - Jun 2021 (6 months)
  • P
    M.S. in Computer
    Pontifical Catholic University of Rio de Janeiro
    Mar 2020 - Sep 2022 (2 years 7 months)
  • DataCamp
    Supervised Learning with scikit-learn
    DataCamp
    Jan 2020 - Sep 2020 (9 months)
  • Coursera
    Natural Language Processing Specialization
    Coursera
    Jan 2020 - Nov 2020 (11 months)
  • P
    B.S.E
    PONTIFICAL CATHOLIC UNIVERSITY OF
    Mar 2015 - Dec 2019 (4 years 10 months)
  • EXIN
    Green IT Citzen
    EXIN
    Jan 2013 - Nov 2013 (11 months)
  • U
    Undergraduate
  • DataCamp
    Linear Classifiers in Python
    DataCamp
Projects (professional or personal) verified_user 0% verified
  • A
    An End-to-End Model for Joint Entity and Relation Extraction in Por tuguese.
Publications verified_user 0% verified
  • a
    aiXplain SDK: A High-Level and Standardized Toolkit for AI Assets
    Jan 2024
  • A
    A Brazilian Portuguese Dataset for Named Entity Recognition and Relation Extraction in the Diabetes Healthcare Domain
    Jan 2023
  • D
    Debugging Machine Translation Systems Based on Their Output
    Jan 2022
  • a
    aiXplain at Arabic Hate Speech 2022: An Ensemble Based Approach to Detecting Offensive Tweets
    Jan 2022
  • a
    at eHealth-KD Challenge 2021: A Multilingual BERT based System for Joint Entity Recognition and Relation Extraction
    Jan 2021