Hi, I'm Mauricio González

BUILDING

PRODUCTS.

Full-Stack Engineer with 5 years of experience building scalable, AI-powered applications. React, Next.js, TypeScript, Node.js, Python — leveraging AI to ship faster.

ABOUT ME ×ABOUT ME ×ABOUT ME ×ABOUT ME ×ABOUT ME ×ABOUT ME ×ABOUT ME ×ABOUT ME ×ABOUT ME ×ABOUT ME ×ABOUT ME ×ABOUT ME ×ABOUT ME ×ABOUT ME ×ABOUT ME ×ABOUT ME ×

ABOUT

ME

I'm a Full-Stack Engineer with 5 years of experience building scalable, AI-powered applications. I've worked across industries — energy, logistics, and SaaS — integrating AI capabilities into real products through APIs, automation, and intelligent features.

I work across the full stack: clean, intuitive interfaces with React and Next.js on the front, robust APIs with Python, Django, and Node.js on the back. I leverage AI tools to accelerate development, optimize workflows, and deliver high-quality products faster.

EXPERIENCE ×EXPERIENCE ×EXPERIENCE ×EXPERIENCE ×EXPERIENCE ×EXPERIENCE ×EXPERIENCE ×EXPERIENCE ×EXPERIENCE ×EXPERIENCE ×EXPERIENCE ×EXPERIENCE ×EXPERIENCE ×EXPERIENCE ×EXPERIENCE ×EXPERIENCE ×

EXPERIENCE

TIMELINE
  • Designed and built a full SaaS product solo — from architecture and database schema to UI and payment integration.
  • Dropshippers generate AI-powered landing pages in minutes, with built-in visitor tracking and automatic optimization suggestions via Claude API.
  • Implemented append-only versioning so every iteration of a landing page is preserved — full history, zero data loss.
  • Built lightweight fire-and-forget behavioral tracking that never blocks page load on the public-facing landing.
  • Chose Lemon Squeezy as Merchant of Record over Stripe to handle global tax compliance and LATAM payment restrictions.
Next.js 14TypeScriptPrismaPostgreSQLClerkAnthropic Claude APICloudinaryRechartsLemon SqueezyVercelSupabase
  • Build and maintain scalable end-to-end systems using React, Next.js, and Node.js.
  • Integrate AI capabilities into real-world products through APIs, automation, and intelligent features.
  • Contribute to system design and architecture decisions for client projects.
  • Improve development workflows by leveraging AI tools across the full stack.
ReactNext.jsTypeScriptNode.jsPostgreSQLDocker
  • Integrated AI-powered features into enterprise energy platforms using React, Redux, and TypeScript, enhancing data analysis and user decision-making.
  • Developed Django-based APIs to orchestrate AI workflows, integrating external AI services and managing data pipelines with PostgreSQL.
  • Implemented asynchronous processing with Celery to handle AI-driven simulations, predictions, and long-running analytical tasks.
  • Enabled real-time feedback and progress tracking for AI processes using WebSockets, improving transparency and user experience.
ReactReduxTypeScriptPythonDjangoWebSocketsDockerPostgreSQLCelery
  • Built a web platform for a logistics company, from UI design in Figma to frontend-backend integration.
  • Implemented an authentication system in Python for secure user access and session management.
  • Developed responsive layouts, animations, and clean UI components with Vue.js.
  • Integrated frontend with backend REST APIs for delivery configuration, payment, and order tracking.
VueJavaScriptPythonCSS/SCSSFigmaREST API
WHAT I BUILD ×WHAT I BUILD ×WHAT I BUILD ×WHAT I BUILD ×WHAT I BUILD ×WHAT I BUILD ×WHAT I BUILD ×WHAT I BUILD ×WHAT I BUILD ×WHAT I BUILD ×WHAT I BUILD ×WHAT I BUILD ×WHAT I BUILD ×WHAT I BUILD ×WHAT I BUILD ×WHAT I BUILD ×

WHAT I

BUILD
01

Platforms & Dashboards

I build internal tools and web platforms that help teams make faster decisions. From energy analytics dashboards to logistics management systems — interfaces designed for clarity and speed.

02

APIs & Backend Systems

I design and build the backend that powers products. REST APIs, data pipelines, async processing, and real-time communication — architected to scale without breaking.

03

Enterprise & Industry Tools

I've worked in energy and logistics, building software that replaces manual processes with structured digital workflows. Systems that handle complexity so users don't have to.

04

End-to-End Product Thinking

I don't just write code — I think about the user flow, the data model, and the long-term maintainability. Clean UI on the front, solid architecture on the back, and real-time when it matters.

PLAYGROUND ×PLAYGROUND ×PLAYGROUND ×PLAYGROUND ×PLAYGROUND ×PLAYGROUND ×PLAYGROUND ×PLAYGROUND ×PLAYGROUND ×PLAYGROUND ×PLAYGROUND ×PLAYGROUND ×PLAYGROUND ×PLAYGROUND ×PLAYGROUND ×PLAYGROUND ×

PLAYGROUND

SIDE PROJECTS

Dropanalizer

Personal SaaS for dropshippers. AI generates landing pages in minutes, tracks visitor behavior, and automatically suggests optimizations. End-to-end product built solo from architecture to payments.

  • Designed and built full SaaS stack alone — from database schema to AI pipeline to payment integration
  • Implemented append-only versioning to preserve full iteration history with zero data loss
  • Built lightweight fire-and-forget behavioral tracking that never blocks public page load
  • Chose Lemon Squeezy as Merchant of Record over Stripe for global tax compliance and LATAM payment support
Next.js 14TypeScriptPrismaPostgreSQLClerkAnthropic Claude APICloudinaryRechartsLemon SqueezyVercelSupabase

Full-stack AI shopping assistant. Handles product search, real-time currency conversion, and intelligent responses using OpenAI Function Calling. Showcases clean architecture with Next.js frontend and NestJS backend.

  • OpenAI GPT-4 with Function Calling for intelligent tool selection
  • Product search from CSV database with smart filtering
  • Real-time currency conversion via Open Exchange Rates API
  • Automatic API documentation with Swagger
  • Docker containerization and CI/CD with GitHub Actions
Next.jsNestJSTypeScriptOpenAI GPT-4DockerSwaggerPostgreSQLGitHub ActionsRailway
AI AGENT ×AI AGENT ×AI AGENT ×AI AGENT ×AI AGENT ×AI AGENT ×AI AGENT ×AI AGENT ×AI AGENT ×AI AGENT ×AI AGENT ×AI AGENT ×AI AGENT ×AI AGENT ×AI AGENT ×AI AGENT ×

AI AGENT

DOCS

This portfolio includes a real AI agent built with the Anthropic Claude API and Next.js Tool Use. It is not a chatbot — it reasons, calls functions, and responds based on live data.

01

What makes it an agent?

A regular chatbot answers from its training data alone. An agent has tools — functions it can call to fetch real information. Claude decides autonomously which tool to use, executes it, reads the result, and then responds. This is the agentic loop.

02 — Agentic loop — step by step
01

User sends a message

A recruiter or visitor types a question in the chat.

02

Claude reasons

The model reads the message and decides which tool — if any — it needs to call.

03

Tool executes

The API route runs real code: filters experience.ts, projects.ts, or returns availability data.

04

Result returns to Claude

The tool output is injected back into the conversation as context.

05

Claude responds

Claude formulates a final answer grounded in real data — not hallucinated.

03 — Available tools
get_experience
Triggered when: "Have you worked with Django?" / "What did you do at Enersinc?"Returns: Filtered work history from experience.ts
get_projects
Triggered when: "Show me your AI projects" / "Do you have SaaS experience?"Returns: Matching projects from projects.ts
get_availability
Triggered when: "Are you available?" / "How do I hire you?"Returns: Availability status + LinkedIn URL
get_stack
Triggered when: "What's your stack?" / "Do you know Next.js?"Returns: Full deduplicated tech stack
04 — Technical stack
LLMClaude Haiku (Anthropic)
Tool UseAnthropic Tool Use API
BackendNext.js App Router API Route
Data sourceexperience.ts · projects.ts · personal.ts
Loopwhile(stop_reason !== end_turn)
05 — Try it

Open the agent (bottom right ↘) and ask one of these:

Have you worked with AI agents?
What projects involve the Claude API?
Are you available for freelance?
What's your full stack?

CONTACT

ME

Open to freelance projects and full-time opportunities. The best way to reach me is through LinkedIn.

Message me on LinkedIn →

Colombia