Skip to content
View eSVeeF's full-sized avatar

Achievements

Achievement: Pull Sharkx2

Achievements

Achievement: Pull Sharkx2

Block or report eSVeeF

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
eSVeeF/README.md

Sergio Vizcaino Ferrer

ML Engineer · Shipping AI/ML systems for production · Contributor to Hugging Face PEFT & Sentence Transformers · Exploring Agentic AI

What I Do

I build AI/ML systems end-to-end — from raw data to production operations.

  • End-to-end delivery — Take AI/ML systems from design and data preparation through deployment, validation, and daily operation.
  • Inference that works in practice — Combine local quantized LLM/VLM inference with cloud services to automate real business workflows.
  • Reliable by design — Add pipelines, monitoring, and recovery workflows so models hold up beyond the demo.
  • Open Source — Shipping fixes to widely used AI/ML ecosystems.

Open Source Contributions

Project Org Merged Reported Focus
Parameter-Efficient Fine-Tuning (PEFT) stars Hugging Face merged PRs reported issues LoRA, gradient scaling, adapter functionality
Sentence Transformers stars Hugging Face merged PRs reported issues Retrieval, embeddings, model validation, inference behaviour
Highlights

Hugging Face · PEFT

  • #3427 — Handled grouped Conv2d rank padding during LoRA hotswapping
  • #3315 — Respected layer-specific scaling in the LoRA-FA optimizer

Hugging Face · Sentence Transformers

  • #3910 — Kept CSR inference from updating dead-feature statistics
  • #3883 — Aligned RRF scoring with canonical 1-based ranking

Tech Stack

Core & ML

Python PyTorch Scikit-learn llama.cpp

Infrastructure & Deployment

Docker AWS CI/CD PySpark

Databases

PostgreSQL MySQL MongoDB Neo4j

AI Platforms

OpenAI Claude GoogleGemini Hugging Face Ollama

Pinned Loading

  1. huggingface/peft huggingface/peft Public

    🤗 PEFT: State-of-the-art Parameter-Efficient Fine-Tuning.

    Python 21.8k 2.6k

  2. huggingface/sentence-transformers huggingface/sentence-transformers Public

    State-of-the-Art Embeddings, Retrieval, and Reranking

    Python 19.2k 3k