Pere Martra
Book Author at Pere Martra
Barcelona, Catalonia, Spain
Summary
🤩
Rockstar🎓
Top SchoolPere Martra is an ML research engineer and author based in Barcelona with 11 years of experience building efficient, secure, and production-ready AI systems. He specializes in LLM post-training optimization—structural pruning, knowledge distillation, and adaptive inference—and is the creator of OptiPfair, an open-source toolkit for activation-guided, fairness-aware pruning that helped his projects retain ~98% performance at half the model size. Author of a forthcoming Manning guide on rearchitecting LLMs, Pere combines hands-on deployment expertise (FastAPI, Docker, CI/CD) with academic rigor from a master’s in AI research and a track record of bias-mitigation validated on models like Llama-3.2 and Salamandra-2B. His background in ATM security and large-scale banking systems informs a pragmatic focus on robustness and responsible AI, and his Large Language Model Notebooks have attracted significant community attention (1.8K+ GitHub stars).
11 years of coding experience
10 years of employment as a software developer
Master's Degree in Artificial Intelligence Research, Artificial Intelligence, Master's Degree in Artificial Intelligence Research, Artificial Intelligence at Universidad Internacional Menéndez PelayoUOC (Universitat Oberta de Catalunya)
English, Spanish, Catalan
Stackoverflow
36reputation
3kreached
4answers
2questions
Github Skills (27)
transformers10
langchain10
vector-database10
large-language-models10
pruning10
huggingface10
model-optimization9
ai9
quantization8
runnable8
bias-detection8
fine-tuning7
language-model7
file-upload6
google-app-engine6
Programming languages (2)
Github contributions (5)
peremartra/optipfair
Apr 2025 - May 2026
Structured pruning, knowledge distillation, and fairness analysis for Large Language Models.
Contributions:5 releases, 7 PRs, 79 pushes in 1 year 1 month
large-language-modelsbias-detectionllmstructured-pruningtransformers
peremartra/Large-Language-Model-Notebooks-Course
May 2023 - May 2026
Practical course about Large Language Models.
Contributions:4 reviews, 14 PRs, 672 pushes in 3 years
large-language-modelschatbotshfhuggingfacelangchain
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.