Senior AI Engineer · Full-stack Python (FastAPI + React) · ML & simulation platforms Remote from Medellín, Colombia (UTC-5, full US Eastern overlap).

I take ML models, LLM pipelines and simulation platforms from prototype to production. On contract to Hoare Lea, a UK engineering consultancy, I took its ML inference services to production (internal users grew from 5 to 50+ engineers) and I'm architecting its cloud platform for long-running physics simulations. 7 years of professional Python, with a background in building-performance simulation.

pabloarango.dev · LinkedIn · CV (PDF)

• Cloud Simulation Job Platform: an Azure job platform for long-running physics simulations. FastAPI job API, queue-triggered workers that scale to zero, React + TypeScript front end. (Hoare Lea; case study only, no code.)
• Graph neural network for overheating prediction: TM59 and Part O compliance in under 10 seconds, about 100x faster than the simulation it replaces. In production use. (Hoare Lea; case study only.)
• Property Valuation Engine: fair asking price per m² for homes listed for sale in the Medellín metro area, with a calibrated interval and the comparable listings behind it. LightGBM, PostgreSQL/PostGIS, GitHub Actions.
• LiDAR2Building API: one latitude/longitude in England in, a 3D site-context model out. FastAPI job service with a Redis queue and PostgreSQL.
• Vision-LLM Event Extraction Pipeline: a scheduled pipeline that reads image posts with a vision model into strict JSON and publishes a searchable events site.

Python · FastAPI · React · TypeScript · PostgreSQL / PostGIS · Docker · Azure · PyTorch · LightGBM · LLM APIs · GitHub Actions