Javier Martinez
Oviedo, Asturias, Spain
Summary
🤩
Rockstar🎓
Top SchoolJavier Martinez is an AI Solutions Architect and full-stack engineer with 10 years of experience building production-grade, resilient AI systems from prototype to scale. Based in Oviedo, Spain, he designs provider-agnostic end-to-end architectures (Hexagonal) and has engineered high-performance hybrid RAG search, semantic model routing, distributed circuit-breakers, and self-correcting learning loops for real-world deployments. He couples strong DevOps and FinOps sensibilities—CI/CD, mutation testing, and SOC2-minded PII scrubbing—with hands-on implementation, contributing across frontend and backend code (notably to a private-GPT project integrating Ollama/Hugging Face models and summarization features). Previously he founded and scaled EdTech products with data-driven growth and automation, and earlier roles cover RPA-driven logistics automation and full-stack .NET/web development. Known for pragmatic architectural rigor, he blends deep systems thinking with an appetite for refining ML systems in production.
10 years of coding experience
2 years of employment as a software developer
Computing Engineering, Ingeniería informática, Computing Engineering, Ingeniería informática at Universitat d'AlacantMaster of Computer Applications (MCA), Informática, Master of Computer Applications (MCA), Informática at Universidad de Murcia
Spanish, English, Spanish
Github Skills (22)
summarization10
python10
ai10
ollama10
llama-index10
back-end-development9
configuration-management9
huggingface9
backend9
gradio9
uid8
front-end-development8
vector-database8
ui-design8
postgresql7
Programming languages (7)
Github contributions (5)
zylon-ai/private-gpt
Mar 2024 - Feb 2025
Interact with your documents using the power of GPT, 100% privately, no data leaks
Role in this project:
Full-stack Developer
Full-stack DeveloperContributions:64 reviews, 57 PRs, 55 pushes in 10 months
Contributions summary:Javier contributed to both the backend and frontend of the project, addressing a variety of issues and implementing new features. They made changes to the vector store implementation, updated configuration settings, and added functionality to the UI, indicating a broad understanding of the project's architecture. A significant portion of their work involved integrating and configuring various LLM models, including Ollama and Hugging Face models, as well as adding a summarization feature. They also updated dependencies and refactored code for improved maintainability.
jaluma/BlackPidgeon
Oct 2019 - Nov 2019
Contributions:37 commits in 1 month
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.