> Markdown version of [/@miguel-martinez](https://www.wearedevelopers.com/@miguel-martinez). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md).

---

# Miguel Martínez
Deep learning data scientist at NVIDIA building efficient and inclusive LLMs.

## About

Miguel Martínez builds deep learning systems at NVIDIA, focusing on generative AI, recommender systems, and data analytics. Earlier in his career, he worked on financial services infrastructure and mentored students through Udacity’s Artificial Intelligence program.

His work today tackles how to make large language models more efficient and globally accessible. He helps developers customize LLMs using parameter-efficient fine-tuning and retrieval-augmented generation. He also creates GPU-accelerated methods for curating multilingual text corpora, solving the data scarcity issues that often hold back non-English model training.



## Past Sessions

### World Congress 2025 · July 9, 2025
Berlin, Germany
- [NVIDIA Expert Session: Accelerating Data Analytics Workflows on GPU Systems](https://www.wearedevelopers.com/events/world-congress-2025/sessions/753-nvidia-expert) · 60 min
- [Breaking Language Barriers: GPU-Accelerated Corpus Curation for LLM Training/Customization](https://www.wearedevelopers.com/events/world-congress-2025/sessions/812-breaking-language) · 120 min
- [NVIDIA Expert Session: Customize Generative AI Models for Industry-Specific Solutions](https://www.wearedevelopers.com/events/world-congress-2025/sessions/897-nvidia-expert) · 60 min

### World Congress 2024 · July 17, 2024
Berlin, Germany
- [Efficient Large Language Model Customization with NVIDIA NeMo Framework.](https://www.wearedevelopers.com/events/world-congress-2024/sessions/170-efficient-large) · 30 min




## Links

- [X](https://twitter.com/miguelusque)
- [LinkedIn](https://www.linkedin.com/in/miguelusque/)