- Built an internal LLM-powered compliance chatbot that quizzes employees on global banking regulations — validating free-text answers and explaining topics on demand.
- Own the service end-to-end: prompt design, answer evaluation, Airflow data flows, Slack integration and Kubernetes deployment.
001 Senior Fullstack Engineer
Luca Tabone
I'm a fullstack engineer in Berlin with nine years across fintech, e-commerce, health-tech and venture capital. I design and own systems end-to-end — from LLM-powered tooling to data backbones that quietly save seven figures a year.
-
0
Years shipping
production software -
€0
Saved per year
through automation -
0
Engineers hired
& led as lead
002 — Selected Work
A decade of shipping, end‑to‑end.
Nine years, seven teams — from founding a startup to leading engineering at a venture fund and building LLM tooling inside a bank. A few that mattered.
- Led backend and data-pipeline development for a German health-tech stealth startup — async Python services from architecture to production.
- Worked cross-functionally with product, design and DevOps to keep delivery aligned across the project's lifetime.
- Led engineering for EagleEye, the fund's deal-sourcing platform — hired 7 engineers, ran sprint and OKR planning, and managed partner-level stakeholders.
- Migrated a multi-repo Node.js / GraphQL / React stack to a t3 monorepo (Next.js, tRPC, Prisma), cutting iteration time and the team size needed to ship.
- Built the data backbone: Airflow / Cloud Composer and Kafka pipelines plus a mobile-proxy crawling system for social platforms.
- Automated exchange and refund handling end-to-end — replacing manual third-party processing and saving €1.5M+ per year.
- Shipped 22 data pipelines (BigQuery, Airflow, Kafka): C-level ad-spend reporting, shipping-status emails, catalog and discount automation.
- Built a React / Node.js stock-management app for the inbound logistics team.
- Founded a digital-loyalty startup: two React Native apps (merchants + customers) with card payments and Apple / Google Pay top-ups.
- Built the serverless Flask API on Cloud Run with Postgres and Firebase Realtime DB for instant in-shop stamp feedback.
- Consulted an automotive client's finance department on Looker data visualisation and contributed to their Java payments backend.
- Built a hierarchical text classifier (142 interconnected MLPs) reaching 96.7% accuracy, trained on data from a self-built Selenium scraping system spanning 10 sources.
003 — Toolkit
The stack I reach for.
Python & TypeScript native, comfortable owning everything from the data layer to the UI and the infrastructure it all runs on.
01 Languages
- Python
- TypeScript
- JavaScript
- SQL
02 Backend & Data
- GraphQL
- REST
- tRPC
- Kafka
- PostgreSQL
- FastAPI
- Node.js
- MongoDB
- BigQuery
- RabbitMQ
- WebSockets
- Pandas
03 Frontend
- React
- Next.js
- Nuxt
- React Native
04 Cloud & Infra
- AWS
- GCP
- Docker
- CI/CD
- Kubernetes
- Terraform
- Serverless
004 — About
I like the unglamorous parts — the pipeline that never breaks, the migration that shrinks the team, the automation that saves a fortune while no one notices. After nine years and seven teams, I've learned that the best engineering tends to be invisible.
- Education
- M.Sc. Computer Science RWTH Aachen University · 2015–2017
- B.Sc. Computer Science RWTH Aachen University · 2012–2015
- Languages
- German Native speaker
- English Native speaker
005 — Contact
Let's build
something that lasts.
Have a project, a role, or a knotty problem worth solving? Tell me a little about it and I'll get back to you within a day or two.
- Based in Berlin, Germany