Skip to content
View edujbarrios's full-sized avatar
🏠
Working from home
🏠
Working from home

Block or report edujbarrios

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
edujbarrios/README.md

Hi, I'm Eduardo — Ed for short 👋

Software Engineer · MSc in AI

Researching and building practical, open-source AI systems.

AI Systems Evaluation & Observability Developer Tools Open Source


🚧 Active Research & Development

TrainLens: Jupyter-native analysis and diagnostics for model-training runs

A Jupyter-first toolkit that reads the metrics and model objects already in your notebook to compare experiments, detect common training problems, and turn run evidence into reproducible reports.

TrainLens downloads

LastLight: low-power, auditable retrieval for disaster and infrastructure-failure scenarios

A stdlib-only Python library that searches local Markdown/ZIP knowledge packs, exposes source passages and ranking metadata, adapts retrieval to resource constraints, and refuses when the available evidence is too weak.

LastLight downloads

📦 Open-Source AI Tooling

Python Package Portfolio

Small, practical Python tools for AI evaluation, data quality, retrieval, multimodal analysis, visualization, and generative-audio workflows.

Retrieval, Context & Agent Evidence

Package Research / engineering focus Downloads
context-dedup Context deduplication for reducing redundant content in LLM and agent workflows. context-dedup downloads
evidenceflow Deterministic analysis of how retrieved and tool-generated evidence propagates through AI-agent traces. evidenceflow downloads
rag-chunk-audit RAG chunk-quality auditing for segmentation and retrieval analysis. rag-chunk-audit downloads

Evaluation, Regression & Safety

Package Research / engineering focus Downloads
embedding-drift-lite Embedding-distribution drift detection and diagnostic inspection. embedding-drift-lite downloads
parametricbench Provider-independent regression benchmarking for LLM and VLM systems. parametricbench downloads
promptshield-llm Prompt-injection, unsafe-instruction, and input-risk screening for LLM pipelines. promptshield-llm downloads

Multimodal & VLM Analysis

Package Research / engineering focus Downloads
metaclean-vlm Metadata normalization for image and vision-language model datasets. metaclean-vlm downloads
visual-patch-audit Image-patch auditing and visual-signal inspection for VLM workflows. visual-patch-audit downloads
vlm-occlusion Grid-based black-box occlusion analysis for VLM claim sensitivity. vlm-occlusion downloads
vlm-prior-probe Counterfactual black-box evaluation of whether VLMs follow visual evidence or learned priors. vlm-prior-probe downloads

Visualization & Developer Experience

Package Research / engineering focus Downloads
matplotlib-dark Automatic dark mode for Matplotlib with ready-to-use themes and safe temporary styling. matplotlib-dark downloads

Generative Audio

Package Research / engineering focus Downloads
text-to-music-prompt-structurer Structured prompt extraction for text-to-music generation workflows. text-to-music-prompt-structurer downloads

🧠 AI Research

Explore Suffice, token-efficiency research for AI agents

Suffice measures how few tokens an AI agent needs to successfully complete a task across prompts, context, tools, memory, and final responses.

It searches for minimum successful token budgets and token-efficiency frontiers under explicit quality constraints, so reducing tokens only counts when task success is preserved.

Suffice token efficiency research

Check MAVERICK, cognition-inspired AI research for multi-agent visual reasoning

Inspired by human visual cognition, MAVERICK decomposes image understanding into an auditable four-agent loop: perceive, describe, critique, and refine.

Designed for interpretable VLM workflows, uncertainty-aware reasoning, and stronger model-ready image descriptions.

Maverick OSS multi-agent VLM

Explore Neural Audio Theory, the engineering foundations behind AI music generation

An open educational project explaining how modern AI music systems work, from signal processing and embeddings to transformer and diffusion architectures, training, and prompt conditioning.

Built for developers and researchers who want a technical understanding of the systems behind neural audio and music generation.

Neural Audio Theory live guide

Discover music-to-text, local-first AI tooling for turning audio into structured metadata and music industry copy

An open-source Python framework that extracts acoustic features and generates A&R notes, PR pitches, playlist descriptions, and sync licensing blurbs, with or without an LLM.

Built for reproducible audio analysis, structured exports, and OpenAI-compatible or fully local workflows.

music-to-text audio intelligence

🧩 UI Components

edujbarrios-ui is a frontend component library for building clean AI interfaces, demos, and developer tools.

AI UI component library

💻 Tech Stack

Languages

Python R Julia Java C C++ Rust

TypeScript JavaScript HTML5 CSS3

AI Systems & Tooling

PyTorch Jupyter LlamaIndex

PostgreSQL pgvector ChromaDB

React Streamlit Gradio Docker

Quarto Docusaurus MDX Jinja


📝 Latest Blog Posts


🎧 Beyond the Code

Creative Side

Music Producer with 7+ years of experience and global reach.

🎵 Listen


🖥 Environment

OS & Development

Windows macOS VSCode Git GitHub

Popular repositories Loading

  1. llm-ts-api-wrapper llm-ts-api-wrapper Public

    A TypeScript wrapper for any OpenAI-compatible APIs — llm7.io, OpenAI, Azure OpenAI, Groq, Together AI, and more.

    TypeScript 3

  2. llm7.io_GUI_tool llm7.io_GUI_tool Public

    A self developed and unofficial frontend to interact with llm7.io API

    Python 2 1

  3. llm7R llm7R Public

    Use powerful Large Language Models (LLMs) directly in your R workflows for data analysis, visualization, and exploratory research.

    R 2

  4. music_creator_agent music_creator_agent Public

    AI assisted music creation using LLM7.io powered agents to generate Suno AI-ready prompts.

    Python 2

  5. plantilla_tfm_muia_uax plantilla_tfm_muia_uax Public template

    Desarrollo desde cero del template en LaTeX para la realización del Trabajo Fin de Máster del MU en Inteligencia Artificial de la Universidad Alfonso X el Sabio (UAX)

    TeX 2

  6. neural-audio-theory neural-audio-theory Public

    An open educational reference connecting practical music-generation workflows with the engineering foundations behind AI music.

    TypeScript 2