Skip to content
View PoeMadara's full-sized avatar
:accessibility:
Looking for job
:accessibility:
Looking for job

Block or report PoeMadara

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.

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
PoeMadara/README.md

Carlos Vergara Gámez

AI Systems Engineer • LLMs • Data Engineering • Edge AI

Designing intelligent systems that combine Artificial Intelligence, Data Engineering, Cloud Infrastructure and Embedded Computing to solve real-world problems.


About Me

I'm an AI & Data Engineer focused on building intelligent systems from data acquisition to deployment.

My work combines Python, Large Language Models, Retrieval-Augmented Generation (RAG), Data Engineering, Cloud Computing, and Embedded Systems to develop scalable AI applications capable of operating in real-world environments.

Rather than specializing in a single layer of the stack, I enjoy connecting them together—from data pipelines and LLM orchestration to edge devices communicating with cloud services.

Outside of my professional work, I enjoy experimenting with emerging AI technologies, geospatial analysis and automation, always looking for practical applications with measurable impact.


Current Focus

  • Large Language Models (LLMs)
  • Retrieval-Augmented Generation (RAG)
  • Multi-Agent Systems
  • AI Evaluation & Prompt Engineering
  • Python Backend Development
  • Data Engineering Pipelines
  • Edge AI & Embedded Systems
  • AWS Cloud Architecture

Professional Experience

AI Engineer (Private Project)

Designed production-oriented AI solutions using hybrid retrieval architectures combining structured databases and semantic vector search.

Worked on LLM orchestration, prompt engineering, FastAPI services and real-time response pipelines.


Embedded Systems Engineer — AURA (Private Project)

Developed firmware for ESP32 wearable devices running FreeRTOS.

Built telemetry pipelines connecting embedded sensors with cloud-based AI services through MQTT and REST APIs.


Data Engineer — GM Quality Technology

Developed large-scale web scraping systems for protected platforms.

Designed ETL pipelines with PySpark and AWS to automate processing of high-volume datasets used by downstream AI workflows.


Featured Projects

3D Multi-Agent AI Platform

Experimental platform exploring multi-agent orchestration, Retrieval-Augmented Generation and tool calling through an interactive 3D interface.

Video Demo 3D Multi-Agent Assistant

Highlights

  • Multi-agent architecture
  • LangChain orchestration
  • RAG pipelines
  • Local and cloud LLM integration
  • Three.js visualization
  • Flask backend

Tech

Python LangChain Flask Vector Databases Three.js AWS


Sustainable Energy Analysis Platform

Machine Learning platform for renewable energy forecasting using statistical analysis and deep learning models.

Built complete data pipelines from preprocessing to model training and visualization.

SEAP Quick Overview GIF

Highlights

  • End-to-end ML pipeline
  • Forecasting models
  • Feature engineering
  • Model evaluation
  • Interactive visualizations

Tech

Python TensorFlow Scikit-Learn Pandas NumPy


Sentinel Fire Risk Prevention

Geospatial analysis platform for early wildfire risk assessment using Sentinel-2 multispectral satellite imagery.

The system automatically retrieves satellite data, processes spectral bands, removes cloud artifacts and calculates vegetation and moisture indices to support environmental monitoring.

Processing Pipeline

System Overview

Input
│
├─ Area Selection
├─ Date Range
└─ Satellite Sources
          │
          ▼
Data Acquisition
          │
          ▼
Geospatial Processing
          │
          ▼
Vegetation Analysis
          │
          ▼
Risk Assessment
          │
          ▼
Maps & Reports

Highlights

  • Sentinel-2 image processing
  • Remote sensing
  • Geospatial analysis
  • Automated satellite downloads
  • Environmental risk assessment

Tech

Python Rasterio NumPy Copernicus STAC Remote Sensing


ATS Resume Generation Pipeline

Deterministic resume compiler that generates ATS-friendly PDFs from structured YAML definitions.

Built to guarantee reproducible document generation, consistent layouts and optimal parsing by Applicant Tracking Systems.

  resume.yaml
        │
        ▼
 Data Validation
        │
        ▼
 HTML Generation
    (Jinja2)
        │
        ▼
 PDF Rendering
  (Playwright)
        │
        ▼
 ATS_Resume.pdf

Highlights

  • YAML-based resume definition
  • Deterministic document generation
  • HTML rendering
  • PDF automation
  • ATS optimization

Tech

Python Jinja2 Playwright YAML


Technical Expertise

Artificial Intelligence

  • Python
  • LLM Applications
  • Prompt Engineering
  • Retrieval-Augmented Generation (RAG)
  • LangChain
  • Hugging Face
  • TensorFlow
  • Keras
  • Scikit-Learn

Data Engineering

  • ETL Pipelines
  • PySpark
  • Pandas
  • NumPy
  • SQL
  • PostgreSQL
  • MySQL
  • MongoDB

Backend Development

  • FastAPI
  • Flask
  • REST APIs
  • WebSockets
  • PyTest

Cloud & DevOps

  • AWS
  • Docker
  • Git
  • Linux

Embedded Systems

  • C++
  • ESP32
  • FreeRTOS
  • MQTT
  • IoT

Other Experience

Indie Game Development

Co-Founder of Crazy Devs 8.

Although game development is no longer my primary focus, it strengthened my experience in software architecture, rapid prototyping, AI behaviours, Unity development and collaborative engineering.

My Games Portfolio GIF/Image


Philosophy

I enjoy building software where AI is only one part of a larger system.

Whether the input comes from databases, web platforms, embedded sensors or satellite imagery, my goal is always the same:

Transform complex data into intelligent, reliable and useful systems.


Connect with Me

Pinned Loading

  1. Sustainable-Energy-Analysis-Platform Sustainable-Energy-Analysis-Platform Public

    This repository uses deep learning and machine learning for time series prediction of electricity demand and energy trends. Includes data, maps, and analysis.

    HTML 4 1

  2. startups-machine-learning-team-1 startups-machine-learning-team-1 Public

    Proyecto para predecir el éxito de startups con Machine Learning. Analizamos características clave con Python. Ayuda a inversionistas y emprendedores. Equipo: Carlos y Danny.

    Jupyter Notebook 2

  3. vanguard-ab-test-team-14 vanguard-ab-test-team-14 Public

    Análisis de un test A/B en el sitio web de Vanguard, comparando la interfaz tradicional y una nueva interfaz. Herramientas: Python, Tableau y Trello. Equipo: Carlos Vergara y Oscar Paul Sanchez.

    Jupyter Notebook 1

  4. my-games my-games Public

    All the games I’ve developed using various languages such as Python, Java, C#, and HTML5.

    Java

  5. JoseMi-Sanchez/sql-database_team-7 JoseMi-Sanchez/sql-database_team-7 Public

    HTML 1

  6. nextjs-erp-system nextjs-erp-system Public

    The system is currently hosted on Firebase, but its modular architecture is ready for future deployment on AWS, using EC2 for hosting, S3 for storage, and DynamoDB with Lambda for serverless services.

    TypeScript 3