Skip to content
View samugc's full-sized avatar
2 followers · 2 following

Achievements

Achievement: Pull Sharkx2

Achievements

Achievement: Pull Sharkx2

Block or report samugc

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
samugc/README.md

Samuel García

AI/ML Engineer · Applied AI · Python & C++

Madrid, Spain · Portfolio · LinkedIn · Email

I build and evaluate AI and software systems with a bias toward measurable evidence: retrieval pipelines, optimisation controllers, quantitative experiments and the tests that make their claims defensible.

  • Double BSc in Mathematics and Computer Engineering, University of Granada.
  • MSc in Artificial Intelligence, Universidad Politécnica de Madrid (completed 2026); thesis graded 10/10.
  • Researcher (R1), Computational Intelligence Group, UPM (Feb–Jul 2026).
  • Accepted paper at CAEPIA 2026 / Springer proceedings (ADR-EDA).
  • English C1.

Focus

  • Applied AI: RAG, retrieval evaluation, evidence-grounded generation, financial NLP.
  • Machine learning and optimisation: dynamic multi-objective optimisation, probabilistic models, quantitative research.
  • Software engineering: Python, C++, FastAPI, React, Docker, reproducible experiments, automated tests.

Selected projects

  • ADR-EDA — reliability-aware controller for dynamic multi-objective estimation of distribution algorithms; published benchmark on 14 problems × 15 seeds.
  • FinRAG — local financial-document question answering with inspectable evidence, source attribution and FinanceBench retrieval evaluation.
  • JobLens — deterministic, evidence-grounded CV tailoring from multiple CVs with loopback security and PDF/DOCX export.
  • QuantLab — systematic portfolio experiments with explicit execution, cost and reproducibility assumptions.
  • MarketNLP — financial sentiment classification comparing a TF-IDF baseline with pretrained FinBERT.

More projects and case studies: samuel-garcia.pages.dev

Stack

Python · C++17 · PyTorch · TensorFlow/Keras · scikit-learn · FastAPI · React · Docker · Git · pytest · NumPy · pandas · SciPy

Contact

Open to junior/early-career AI Engineer, Applied ML, Data Scientist and Python Software Engineer roles in Madrid or remote Spain.

Pinned Loading

  1. ADR-EDA ADR-EDA Public

    Adaptive control over persistent estimation of distribution algorithms for dynamic multiobjective optimization.

    Python

  2. FinRAG FinRAG Public

    Local RAG for financial documents with TF-IDF and MiniLM retrieval, Recall@k evaluation, source attribution and local Qwen generation.

    Python

  3. QuantLab QuantLab Public

    Quantitative research framework for systematic portfolio experiments, simulation, risk analysis and reproducible validation.

    Python

  4. SatelliteVision SatelliteVision Public

    Satellite image classification on xView with Swin Transformers, imbalance-aware training, TTA and reproducible evaluation.

    Jupyter Notebook

  5. MarketNLP MarketNLP Public

    Financial sentiment classification comparing TF-IDF + Logistic Regression with pretrained FinBERT under a reproducible evaluation pipeline.

    Python

  6. StochasticDiffusion StochasticDiffusion Public

    C++ parameter estimation for stochastic diffusion models using numerical, evolutionary and memetic optimisation.

    C++