Eugenio Herrera-berg
Lead Machine Learning Engineer at Centro Nacional de Inteligencia Artificial
Santiago, Santiago Metropolitan Region, Chile
Summary
🤩
Rockstar🎓
Top SchoolEugenio Herrera-berg is a Lead Machine Learning Engineer based in Santiago with a decade of experience building end-to-end ML systems, from pre-training multi-billion-parameter LLMs to deploying production CV and NLP services. He currently leads development of LatamGPT (8B–70B models on H200 clusters), virtual try-on pipelines, and benchmarking/synthetic-data tooling while guiding cross-functional research and engineering teams. His work spans infrastructure (distributed training factories, GCP, FastAPI), model research (DPO, CLIP fine-tuning for Spanish), and product-facing latency optimizations that cut inference time by 40%. An adjunct lecturer in generative computer vision, he blends academic rigor—co-authoring multiple papers including ICLR 2026—with hands-on delivery and notable open-source contributions.
10 years of coding experience
3 years of employment as a software developer
Bachelor's degree, Computer Science, Bachelor's degree, Computer Science at Pontificia Universidad Católica de Chile
Spanish, English
Github Skills (111)
transformers10
language-model10
python10
stylegan10
discriminator10
flax10
deep-learning10
transformer10
nlp10
jax10
multimodal-deep-learning10
seq2seq10
tensorflow10
clip10
natural-language-processing10
Programming languages (6)
Github contributions (5)
ouhenio/StyleGAN3-CLIP-notebooks
Oct 2021 - Mar 2022
A collection of Jupyter notebooks to play with NVIDIA's StyleGAN3 and OpenAI's CLIP for a text-based guided image generation.
Contributions:18 commits, 16 PRs, 43 pushes in 5 months
pytorchstyleganpythonnvidiadeep-learning
ouhenio/stylegan2-ada-custom-projector
Oct 2020 - Sep 2021
StyleGAN2 with adaptive discriminator augmentation (ADA) - Official TensorFlow implementation
Contributions:84 commits, 6 PRs, 80 pushes in 10 months
adadeep-learningadaptivestylegan2discriminator
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.