Skip to content
View Pablodeharo's full-sized avatar

Block or report Pablodeharo

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
Pablodeharo/README.md

🔍 Open to AI/ML Engineer & Data Scientist roles — Spain & internationally


I was born in Ghent, Belgium, and moved to Spain at 14. Growing up between two cultures taught me to adapt fast, communicate across contexts, and read what people actually need — skills I now apply to how I build AI systems.

Before moving into AI, I spent years in competitive commercial environments: B2B fleet sales, real estate advisory on the Costa del Sol, and customer-facing roles at Málaga Airport. That background gives me something technical profiles often lack: a real understanding of business impact and user needs.

I trained in Data Science at 4GeeksAcademy and went deep into agentic AI — LangChain, LangGraph, LLM integration, and vector databases. I build systems where intelligence is the core of the product, not just a feature.


🚀 Flagship Projects

Cybersecurity-focused AI agent combining RAG-based retrieval with MCP tool orchestration.

  • Contextual reasoning over security documents
  • Automated incident analysis via MCP tool orchestration
  • Regulatory reasoning aligned with ENS, RGPD, and NIS2 frameworks

Python FastMCP Elasticsearch MCP-Inspector LangChain


A Socratic AI agent trained on the complete works of Plato, designed to guide users through dialectical reasoning instead of giving direct answers.

  • Hybrid RAG pipeline (semantic kNN + BM25) over 3 specialized Elasticsearch indices
  • Orchestrated with LangGraph for stateful multi-node philosophical workflows
  • Decoupled prompt engineering using Markdown templates and Pydantic validation

Python 3.12 LangGraph Elasticsearch Pydantic v2 Poetry Scrapy

🎓 Low-level Programming — Campus 42

Participated in the intensive la piscine selection process and completed core projects applying C programming, memory management, and Linux system administration in a peer-to-peer learning environment.

Project Description
Libft Reimplementation of essential C standard library functions
ft_printf Custom printf with variadic functions and formatted output
get_next_line Efficient file reading using dynamic memory
Born2beroot Linux system administration and security on Debian

🛠 Skills

🧠 AI & Machine Learning

LangChain LangGraph RAG LLM Integration Prompt Engineering TensorFlow PyTorch scikit-learn Elasticsearch ChromaDB

💻 Programming & Data

Python C Pandas NumPy Matplotlib Seaborn

🌐 Web & Deployment

Streamlit Flask Django Docker Git Linux Poetry

☁️ Cloud

AWS Certified Cloud Practitioner


📫 Contact

LinkedIn WhatsApp Gmail

Pinned Loading

  1. Sentiment-Analysis-with-BERT-Light Sentiment-Analysis-with-BERT-Light Public

    Jupyter Notebook

  2. VoltWorth VoltWorth Public

    Python

  3. Voltworth-core Voltworth-core Public

    Jupyter Notebook

  4. llm-relational-db llm-relational-db Public

    Python

  5. Fraude-en-transacciones Fraude-en-transacciones Public

    Jupyter Notebook