Skip to content
View adrmisty's full-sized avatar
🎯
Focusing
🎯
Focusing

Achievements

Achievement: Pull SharkAchievement: Quickdraw

Achievements

Achievement: Pull SharkAchievement: Quickdraw

Highlights

  • Pro

Block or report adrmisty

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
adrmisty/README.md

Adriana R. Flórez

GenAI Engineer @ ArcelorMittal Global R&D | NLP Specialist | Computational Linguist

I build applied AI and language technology, from multilingual NLP research to generative AI solutions in the steel industry. I am a firm believer in tech for social good and ethical AI.

nagare: a year of my GitHub activity drawn as a river

Experience

  • R&D Engineer @ ArcelorMittal Spain (Oct 2026 – Present) | Avilés, Spain
    • IT solutions for ArcelorMittal Global R&D based on state-of-the-art generative AI/vision models.
  • NLP Researcher Intern @ Vicomtech (Nov 2025 – Jun 2026) | Donostia, Spain
    • 🥇 1st place in the SEPLN/IBERLEF 2026 shared task GRACE (evidence-based medicine): argument mining, x-AI and prompt engineering with Qwen and MedGemma.
    • EU-Farmbook (EU R&D): synthetic data pipelines and RAG evaluation with generative AI (Gemini, OpenAI).
    • Document quality validation with multilingual neural topic modeling (BERTopic).
  • Data Scientist Intern @ Bedrock (Mar – Jun 2024) | Gijón, Spain
    • Computer vision models in PyTorch for defect detection in smart manufacturing.
    • B2B retail recommender systems on Azure ML.
  • Front-End Developer Intern @ Transparent Edge (Jun – Aug 2023) | Remote
    • CDN availability dashboards and UI components; database refactoring for REST API integration.

Education

  • MSc in Language and Communication Technologies (Erasmus Mundus LCT) (2024 – 2026)
    • Charles University (Czechia) & UPV/EHU (Spain)
    • Thesis: Automatic detection and adaptation prediction of lexical borrowings in low-resource languages (Asturian, Basque, Greek).
  • BSc in Software Engineering (2020 – 2024)
    • University of Oviedo (Spain), with exchanges at Virginia Tech (USA) and Haliç University (Türkiye)
    • Thesis: LLM-based automatic multilingual localization service for software applications.

Skills

  • GenAI & NLP: LLMs (Gemini, OpenAI, Qwen, MedGemma), RAG, prompt engineering, Hugging Face, spaCy, NLTK, BERTopic, WhisperX, UDPipe
  • ML & Data: PyTorch, TensorFlow, scikit-learn, pandas, NumPy, Azure ML
  • Programming: Python, Java, C++, C#, Swift, JavaScript, HTML/CSS
  • Engineering: Git, Docker, Bash, FastAPI, Linux, MySQL, Oracle
  • Human Languages: Spanish & Asturian (Native), English (C2), Greek (B2), German (B1), French (A2), Basque (A1)

Connect

Pinned Loading

  1. tfm-low-res-lexical-borrowings tfm-low-res-lexical-borrowings Public

    Repo for the development of my Master's thesis (ÚFAL / EHU) on 'Automatic detection and prediction of lexical borrowings in low-resource languages'.

    Python 1

  2. grace-iberlef26 grace-iberlef26 Public

    GRACE 'Granular Recognition of Argumentative Clinical Evidence' shared task work for IBERLEF26, with preliminary work on the CasiMedicos-Arg dataset

    Python 1

  3. doc-quality doc-quality Public

    Project for automatic document quality validation (structural & topical relevance).

    Python 1

  4. deep-learning deep-learning Public

    NPFL138 course projects @ ÚFAL Matfyz (Fall '24), taught by Milan Straka.

    Python 1

  5. synth-qa-RAG-dataset synth-qa-RAG-dataset Public

    Generation of a silver standard Q&A dataset for RAG evaluation.

    Python 1

  6. parallel-speech-data parallel-speech-data Public

    NPFL087 Machine Translation project for achievement of Spanish-Greek parallel audio dataset, taught by Ondřej Bojar.

    Jupyter Notebook 1