Results-driven backend developer with 5+ years building scalable systems and AI-powered applications. Specialized in microservices architecture, distributed systems, and LLM integration.

Languages
Python • JavaScript • Go • SQL • Bash

Backend & AI
FastAPI • Django • Gin • Flask • Celery
LangChain • LangGraph • Strands • DSPy • AWS Bedrock

Data & Storage
PostgreSQL • MySQL • MongoDB • ArangoDB
ChromaDB • Pinecone • Vector Databases

Cloud & Infrastructure
AWS (Lambda, S3, DynamoDB, Bedrock, SQS, ECS) • Azure • Docker • Kubernetes
OpenTelemetry • Grafana • LangFuse • Jaeger • DataDog

Architecture & Practices
Microservices • REST APIs • Event-Driven (RabbitMQ)
TDD • SOLID • CI/CD • Distributed Tracing

• Led enterprise chatbot migration from AWS Bedrock Agents to the Strands framework.
• Designed multi-agent architectures with specialized agents branching by business function.
• Built an LLM-as-Judge evaluation pipeline: an automated self-improvement loop scoring outputs against a Golden Dataset.
• Integrated RAG systems with vector databases for context-aware, highly accurate responses.

• Developed high-performance microservices with FastAPI and Aiohttp.
• Migrated a monolithic system to a microservices architecture (Python 3.11).
• Implemented distributed tracing (Jaeger/OpenTelemetry), reducing MTTR by 40%.

• Integrated 15+ third-party APIs, maintaining 99.9% uptime.
• Optimized complex Django queries, improving search performance by 60%.
• Managed 100GB+/day log pipelines with SumoLogic at scale.

Autonomous research agent for personalized news curation | View Repo

AI-powered platform that creates personalized news briefings from natural language descriptions. Features an autonomous Research Agent that discovers sources, validates content, and keeps you updated.

What it does:

• 🧠 Research Agent v2.0: ReAct-based autonomous agent that plans searches, reasons about relevance, and validates sources
• 🤖 AI Code Reviewer: Internal Gemini-powered agent for automated PR reviews
• 🔍 Multi-LLM Support: Claude for validation, Gemini for extraction
• 📚 Smart Briefings: Auto-curated articles grouped by source with AI-generated descriptions
• 🎯 NL Input: Describe topics in plain language, agent handles the rest

Technical Highlights:

• Autonomous Agent Architecture: ReAct loop with planning, search, and validation steps
• Multi-Search Integration: Tavily API + Google Custom Search + web scraping fallback
• Dual Database Design: PostgreSQL (relational) + DynamoDB (NoSQL)
• Production Infrastructure: AWS CDK with ECS Fargate, RDS, Amplify
• Full-Stack: FastAPI backend + Next.js 14 frontend with glassmorphism UI

Tech Stack:
Python FastAPI Strands Claude SDK Gemini API Next.js 14 TypeScript PostgreSQL DynamoDB AWS CDK ECS Fargate Tailwind CSS

Key Features:

• ⚡ Streaming UI for Research Agent progress visualization
• 🔄 Auto-refresh keeps briefings updated
• 🎨 Modern glassmorphism design with shadcn/ui
• 🧪 Full CI/CD with GitHub Actions + AI PR reviews
• 📱 Mobile-responsive interface

Status: 🚀 Active development • AWS deployment ready

📖 Documentation • 🏗️ Architecture

Intelligent job opportunity analysis and automation | View Repo

Enterprise-grade automation system for LinkedIn job search. Scrapes messages, analyzes opportunities with AI, scores matches, and generates personalized responses.

What it does:

• 📥 Daily LinkedIn message scraping with Playwright
• 🧠 AI-powered analysis using DSPy structured prompting
• 📊 Multi-dimensional scoring and A/B/C/D tier classification
• ✍️ Context-aware response generation adapted to job search status
• 📧 Daily summary emails with all opportunities
• 🎨 React dashboard for opportunity management

Technical Highlights:

• Production Observability: Full monitoring stack (Prometheus, Grafana, Jaeger, Loki)
• Multi-LLM Support: Works with Ollama (local), OpenAI, or Anthropic
• Smart Caching: Redis-based caching reduces LLM calls by 60%
• Async Architecture: FastAPI + Celery + Redis for background jobs
• 85% Test Coverage: 140+ tests including integration and performance

Tech Stack:
Python FastAPI DSPy Playwright React PostgreSQL Redis Celery Docker Prometheus Grafana

Key Features:

• ⚡ Real-time SSE streaming for analysis progress
• 🎯 Pre-configured Grafana dashboards
• 🔄 Complete CI/CD pipeline
• 📱 Mobile-responsive frontend with Tailwind + shadcn/ui
• 🧪 Mailpit integration for email testing

Built as: Technical challenge showcasing enterprise architecture

📖 Full Documentation • 🚀 Quick Start

Comprehensive brewing management platform | In Development

Mobile-first application for homebrewers to manage recipes, track fermentation, and analyze brewing sessions.

Features:

• 📝 Recipe builder with grain bills, hop schedules, and calculations
• 📊 Fermentation tracking with temperature/gravity logging
• 🗄️ 400+ ingredient database (malts, hops, yeasts)
• 📈 Brewing analytics and efficiency tracking
• 📱 Offline-first architecture

Tech Stack:
React Native Expo TypeScript FastAPI PostgreSQL AWS S3

Status: 🚧 Active development • Beta release Q2 2025

🏆 Data Structures & Algorithms - Jovian (2025)
🏆 Systems Design Expert - AlgoExpert (2023)
🏆 MITx: CS & Programming with Python (2022)

💡 Open to senior backend and AI engineering opportunities
🌎 Based in Argentina • Remote-ready • Spanish & English