Backend developer. I build the invisible engine: APIs, queues and services that hold up under real load, in Go, Python and Node.
Backend DevAcapulco, MX# projects
Five working systems. Each one with the problem it solves, how I solved it, and what I measured.
enlarge ⤢High-concurrency ticket sales with zero double-booking, guaranteed under real load
Hexagonal Go with transactional locks in Postgres. Redis releases the reservation after 10 min and RabbitMQ acts as a waiting room.
enlarge ⤢Real-time status monitor for external APIs, with automatic incident detection
An asyncio worker checks every service with no DB credentials; the backend opens and closes incidents and broadcasts them over WebSocket.
enlarge ⤢100% local AI chat: no cloud, no telemetry, text and images in the same conversation
Ollama and Stable Diffusion orchestrated by an intent classifier, streaming over Server-Sent Events.
enlarge ⤢AI-powered candidate screening: scoring, coverage, and explained gaps
An LLM compares each CV to the job and returns a score, coverage and missing skills, with rules so score and reasons agree.
enlarge ⤢Intelligent support system with retrieval-augmented generation
ChromaDB sends the AI only relevant context, Redis caches repeated questions, and NestJS guards the FastAPI pipeline.
# about
I get excited about the cases nobody tests: two people buying the same seat in the same millisecond, a worker dying at 3 a.m., a SELECT that suddenly takes too long. That’s where I like to work.
# contact
Recent graduate, available for remote work as a junior. If you have a backend with interesting problems, I'd love to hear about it.