Albert Luppol

Autonomy Scientist,
Software Engineer.

I build the data platform at Fusion Anesthesia Solutions and study machine learning at Georgia Tech (MSCS). Event-driven intake, per-source parsers, rules that open and close their own work. Python · PyTorch · FastAPI.

What I do

My work is the platform that ingests billing data from fourteen provider organisations — one parser strategy per source so onboarding a client adds files instead of editing the ones that already work, and a rules engine on top that opens a task when a record needs a human and closes it when later data proves the problem is gone.

Underneath that was a migration: a chain of stored procedures became an event-driven flow, where each step reacts to a recorded fact rather than being called by the step before it. You buy a trace of what happened and failure isolation; you pay with transactions you no longer get for free.

Alongside it I am doing an MSCS at Georgia Tech in machine learning — mostly to know precisely which side of the line a problem sits on: the conditions you can write down, and the patterns nobody can.

PythonPyTorchFastAPITypeScriptJavaDjangoSpring BootReactPostgreSQLMongoDBDocker

Selected work

Life Balance

A planner that organises goals and initiatives around long-term values instead of a to-do list. Spring Boot service over Postgres, containerised, with tests and coverage running on every push.

LiveJava 21 · Spring Boot · Postgres

Crazy Fingers

A typing trainer for getting comfortable with a new keyboard — real text, live accuracy, no account and no ceremony. Runs entirely in the browser.

LiveAngular · TypeScript

Nanolink

A URL shortener split the way it would be split in production: a FastAPI service that owns writes, a Fastify service that does nothing but redirect fast, MongoDB between them.

LiveFastAPI · Fastify · MongoDB

Contact

Open to conversations about backend and platform work.

albert.y.luppol@gmail.com · LinkedIn · GitHub