Likes
Comments
Status
Error loading CV
Showing raw YAML instead:
cv:
name: Alexander Oviedo Fadul
location: Sincelejo, Sucre, Colombia
email: aoviedofadul@gmail.com
phone: +57 301 5930519
summary: 'Generative AI / Computer Vision Engineer especializado en imagen y video sintetico con modelos de difusion (Stable Diffusion XL, FLUX, GLM-Image) y multimodalidad vision-language (MiniCPM-V, LLaVA, GPT-4V). Construyo pipelines completos: data curation, training/fine-tuning con LoRA y DreamBooth, deployment con ComfyUI workflows, y orchestration con FastAPI.
Creador de PixelCV, plataforma que genera pixel art con Stable Diffusion fine-tuneado, edicion sprite-by-sprite con tooling propio (pincel, relleno, gotero, undo/redo), galeria comunitaria y OG images dinamicas. Tambien publico modelos en Hugging Face (https://huggingface.co/bladealex) y Ollama (https://ollama.com/bladealex) con cuantizaciones GGUF para inferencia edge.
Stack: PyTorch 2.x, Diffusers (HuggingFace), ComfyUI con custom nodes, ControlNet para guidance espacial, IP-Adapter para style transfer, AnimateDiff para video, OpenCV + Pillow para preprocessing, Stable Diffusion WebUI Forge. Modelos vision-language: MiniCPM-V para OCR estructurado, LLaVA para descripcion contextual, Florence-2 para deteccion + grounding.'
social_networks:
- network: LinkedIn
username: alexander-oviedo-fadul
- network: GitHub
username: alexanderoviedofadul
sections:
Experiencia:
- company: PixelCV (https://pixelcv.dev)
position: Founder & Lead AI Engineer
start_date: 2024-01
end_date: present
location: Remoto
highlights:
- Plataforma open-source para crear CVs con IA + pixel art generado con Stable Diffusion
- 'Stack: Next.js 16 + React 19 + FastAPI + MariaDB + Multi-AI fallback (Groq + DeepSeek + Ollama)'
- Editor de pixel art frame-by-frame con herramientas custom (pincel, relleno, gotero, undo/redo)
- Generacion OG images dinamicas con vercel/og + composicion programatica
- 16 juegos arcade retro con IA (Minimax, Alpha-Beta, A* pathfinding) + sistema de gamificacion
- 471 proyectos GitHub indexados con embeddings semanticos para Projects Explorer
- company: Hugging Face & Ollama Hub
position: Open-Source Model Publisher
start_date: 2023-06
end_date: present
location: Remoto
highlights:
- Publicacion de modelos LLM cuantizados (GGUF) en Hugging Face
- Distribucion de TutelaBot family (3 modelos juridicos colombianos) en Ollama Hub
- 'VulnSentinel: IA de 2.7B parametros para operaciones de ciberseguridad offline'
- Documentacion model cards completas + guias de uso responsable
Educacion:
- institution: ENEB Barcelona
area: Master en Project Management + Master en IA Empresarial
start_date: 2024-01
end_date: 2025-01
location: Barcelona, Esp.
degree: Grado
- institution: Instituto Europeo de Posgrado IEP
area: Master en Business Intelligence y Big Data
start_date: 2021-01
end_date: 2022-01
location: Madrid, Esp.
degree: Grado
- institution: ENEB Barcelona
area: Master en Big Data & Business Intelligence
start_date: 2020-01
end_date: 2021-01
location: Barcelona, Esp.
degree: Grado
- institution: Universidad Isabel I de Castilla
area: Master en Big Data & Business Intelligence
start_date: 2020-01
end_date: 2021-01
location: Burgos, Esp.
degree: Grado
- institution: CORPOSUCRE
area: Derecho
start_date: 2018-02
end_date: 2024-05
location: Sincelejo, CO
degree: Grado
- institution: CUN
area: Ingenieria de Sistemas
start_date: 2012-01
end_date: 2016-01
location: Sincelejo, CO
degree: Grado
- institution: CUN
area: Tecnologo en Soluciones de Software y Redes
start_date: 2010-01
end_date: 2015-01
location: Sincelejo, CO
degree: Grado
- institution: CUN
area: Tecnico Profesional en Ingenieria de Sistemas
start_date: 2005-01
end_date: 2007-01
location: Sincelejo, CO
degree: Grado
- institution: University of Michigan
area: 'IA generativa: impacto en las empresas y la sociedad'
start_date: 2026-04
end_date: 2026-04
- institution: IBM
area: 'IA Generativa: Modelos y Plataformas Fundamentales'
start_date: 2026-04
end_date: 2026-04
- institution: IBM
area: Creacion de aplicaciones generativas basadas en IA con Python
start_date: 2026-03
end_date: 2026-03
- institution: Universitat Autonoma de Barcelona
area: Como persuadir? Jugando con palabras, imagenes y numeros
start_date: 2026-03
end_date: 2026-03
- institution: IBM
area: 'IA Generativa: Introduccion y Aplicaciones'
start_date: 2026-02
end_date: 2026-02
- institution: IBM
area: Construir aplicaciones de IA generativa multimodal
start_date: 2026-02
end_date: 2026-02
Proyectos:
- name: conversacion-claude
highlights:
- Este proyecto es una interfaz de línea de comandos para mantener conversaciones con el asistente de Anthropic llamado Claude. Utiliza la API de Anthropic para enviar mensajes y recibir respuestas del asistente.
- 'Stack: Python, Anthropic, Claude, claude-3'
- 'Codigo: https://github.com/alexanderoviedofadul/conversacion-claude'
- name: IsabellaEA-ai-rag
highlights:
- Este repositorio contiene todo lo necesario para poder crear tu propia secretaria hecha con Inteligencia Artificial, todo gracias al RAG de Isabella AI, Esto fue creado para poder facilitarle la vida a las personas con la IA.
- 'Stack: Python'
- 'Codigo: https://github.com/alexanderoviedofadul/IsabellaEA-ai-rag'
- name: open-webui
highlights:
- User-friendly WebUI for LLMs (Formerly Ollama WebUI)
- 'Demo: https://openwebui.com'
- name: '01'
highlights:
- The open-source language model computer
- 'Demo: http://openinterpreter.com/01'
- name: cody
highlights:
- AI that knows your entire codebase
- 'Demo: https://cody.dev'
- name: chatbot-ollama
highlights:
- Chatbot Ollama is an open source chat UI for Ollama.
- 'Codigo: https://github.com/alexanderoviedofadul/chatbot-ollama'
- name: openui
highlights:
- OpenUI let's you describe UI using your imagination, then see it rendered live.
- 'Codigo: https://github.com/alexanderoviedofadul/openui'
- name: codel
highlights:
- ✨ Fully autonomous AI Agent that can perform complicated tasks and projects using terminal, browser, and editor.
- 'Demo: https://discord.gg/uMaGSHNjzc'
- name: anything-llm
highlights:
- The all-in-one Desktop & Docker AI application with full RAG and AI Agent capabilities.
- 'Demo: https://useanything.com'
- name: Langchain-con-Ollama
highlights:
- Langchain repo with Ollama and Llama 3 - 8b integration for RAG
- 'Codigo: https://github.com/alexanderoviedofadul/Langchain-con-Ollama'
- name: AI-Bootcamp
highlights:
- Self-paced bootcamp on Generative AI. Tutorials on ML fundamentals, LLMs, RAGs & AI Agents
- 'Demo: https://www.mlexpert.io/bootcamp'
- name: langchain-tutorials
highlights:
- A set of LangChain Tutorials from my youtube channel
- 'Codigo: https://github.com/alexanderoviedofadul/langchain-tutorials'
- name: pywinassistant
highlights:
- 'The first open source Large Action Model generalist Artificial Narrow Intelligence that controls completely human user interfaces by only using natural language. PyWinAssistant utilizes Visualization-of-Thought Elicits Spatial Reasoning in '
- 'Codigo: https://github.com/alexanderoviedofadul/pywinassistant'
- name: easy-llama
highlights:
- Text generation in Python, as easy as possible
- 'Codigo: https://github.com/alexanderoviedofadul/easy-llama'
Modelos Publicados:
- name: TutelaBot (3 versiones)
highlights:
- IA pionera para tutelas colombianas
- 'Disponible en Ollama Hub: bladealex/tutelabot'
- name: VulnSentinel
highlights:
- IA tactical 2.7B params (Gemma 4 base)
- VulnSentinel-Pro E4B para SAST/DAST con razonamiento
- name: PixelCV Pixel Art Models
highlights:
- Stable Diffusion fine-tuned para 16x16 / 32x32 sprites
- 'Style: NES/SNES/16-bit retro'
Habilidades:
- label: Diffusion Models
details: Stable Diffusion 1.5/2.1/XL/3, FLUX, GLM-Image, ControlNet (Canny, Depth, Pose, Scribble), IP-Adapter, T2I-Adapter
- label: Fine-Tuning
details: LoRA, DreamBooth, Textual Inversion, Hypernetworks, full fine-tune con 8-bit Adam, mixed precision (fp16/bf16)
- label: Inference Tools
details: ComfyUI (custom nodes propios), Stable Diffusion WebUI Forge, AUTOMATIC1111, Diffusers (HuggingFace), InvokeAI
- label: Video Gen
details: AnimateDiff, Stable Video Diffusion, CogVideoX, ModelScope T2V, frame interpolation con RIFE
- label: Vision-Language
details: MiniCPM-V (OCR estructurado), LLaVA, Florence-2 (deteccion + grounding), GPT-4V, Claude Vision
- label: Computer Vision Clasica
details: OpenCV, Pillow, scikit-image, YOLOv8/YOLOv11, Segment Anything (SAM 2), DeepLab v3+
- label: 3D & Neural Rendering
details: NeRF basics, 3D Gaussian Splatting, Marching Cubes, Open3D, PyTorch3D
- label: Deployment
details: Modelos cuantizados GGUF (Q4_K_M, Q5_K_S, Q8_0), llama.cpp, vLLM, TensorRT, ONNX Runtime, mobile-optimized via TFLite/CoreML
- label: MLOps
details: HuggingFace Hub, Ollama Hub, Weights & Biases, MLflow, DVC para datasets, model cards y dataset cards completos
- label: Stack Python
details: PyTorch 2.x, Transformers, Diffusers, Accelerate, BitsAndBytes, xFormers, Triton kernels, FlashAttention 2
design:
theme: engineeringresumes
locale:
language: spanish