Pablo Sanchez
Software And AI Engineer at DOCYET
Karlsruhe, Baden-Württemberg, Germany
Summary
👤
Senior🎓
Top SchoolPablo Sanchez is a Software and AI Engineer based in Karlsruhe with 8 years of experience building ML and NLP-driven solutions, particularly around LLMs and data modeling. He combines a strong research background—Master’s work at DFKI on clinical relation extraction and active learning—with practical product experience at DOCYET, moving from intern to engineer. Comfortable across the stack, he has implemented production tools (Flask, EmberJS, Rails, Docker) and focuses on interpretable ML, contributing a TokenClassificationExplainer to transformers-interpret that enhances NER explainability using Captum. An active Julia library contributor and functional-programming enthusiast, he pairs rigorous experimentation (PyTorch, HF-Transformers) with test-driven development. Fluent in German and experienced in interdisciplinary collaboration, he brings both academic depth and pragmatic engineering to AI systems.
8 years of coding experience
Bachelor of Engineering - BE, Computer Software Engineering, 8.78/10 in the Spanish system, equivalently 3.3 in 4.0 scoring, Bachelor of Engineering - BE, Computer Software Engineering, 8.78/10 in the Spanish system, equivalently 3.3 in 4.0 scoring at Universidad de ValladolidIntensive German C1 Course, German Language and Literature, Intensive German C1 Course, German Language and Literature at Elefant SprachschuleMaster of Science - MS, Artificial Intelligence, 1.5 in the German system, equivalently 3.5 in 4.0 scoring, Master of Science - MS, Artificial Intelligence, 1.5 in the German system, equivalently 3.5 in 4.0 scoring at Saarland University
Spanish, English, German
Github Skills (13)
transformers10
pytorch10
xai10
machine-learning10
nlp10
explainable-artificial-intelligence10
python10
text-classification10
natural-language-processing10
classification10
classify10
deep-learning9
testing8
Programming languages (10)
Github contributions (5)
cdpierse/transformers-interpret
May 2022 - Jun 2022
Model explainability that works seamlessly with 🤗 transformers. Explain your transformers model in just 2 lines of code.
Role in this project:
ML Engineer
ML EngineerContributions:20 commits, 2 PRs, 3 comments in 1 month
Contributions summary:Pablo primarily contributed to the development of a token classification explainer, a tool designed to enhance the interpretability of transformer models for Named Entity Recognition (NER) tasks. Their work involved creating a `TokenClassificationExplainer` class, implementing methods for encoding and decoding text, and integrating attribution calculations using the Captum library. The user also focused on improving the explainer's functionality by allowing for the specification of ignored indexes and labels, and modifying the data structures for representing and displaying attributions. Furthermore, they developed tests for the explainer to ensure the tool works as expected.
bertnatural-language-processingexplainabilitytransformerslines
pabvald/chatbot2
May 2020 - Mar 2025
**RelintBot2** - Prototype of a chatbot for the service of International Relations of the University of Valladolid (UVa)
Contributions:9 PRs, 16 pushes, 1 branch in 4 years 11 months
relationsbotuvavalladolidprototype
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.