Home Original page

LeaYeh - Overview

Hi, I'm Lea Yeh 👋

Sr. Software Engineer · Data Engineer & MLOps

Profile views

About Me

  • 🏗️ Currently at c-sense GmbH (Vienna) as a Sr. Software Engineer — designing modular, event-driven software architecture for nanoscale scientific instruments, collaborating with hardware engineers and research scientists
  • 📊 8+ years across Data Engineering, Data Science, and Software Engineering — including 7 years at MediaTek building PB-scale data pipelines, ML model deployment, and observability infrastructure
  • 🎓 MSc Computer Science, National Chiao Tung University, Taiwan
  • 🔧 Concurrently enrolled at 42 Vienna, focusing on Linux kernel internals, software architecture (SOLID), and DevOps — targeting a transition into AI Platform / MLOps / SRE roles
  • 📍 Based in Vienna, Austria

What I'm Working On

  • 🐧 ft_linux — Built Linux from scratch: cross-compilation toolchain, custom kernel, bootloader, init system
  • ☸️ Inception-of-Things — K3s multi-node Kubernetes cluster with GitOps/ArgoCD
  • 📐 libftpp — C++20 library applying SOLID principles and hexagonal architecture, with ADRs
  • 🤖 litetorch / litetune / sklite — Neural net framework, hyperparameter tuner, and ML toolkit built from scratch

Technical Skills

Architecture: Event-driven systems · Layered design · SOLID · Hardware abstraction

Data Platform: Airflow · NiFi · Dataflow · BigQuery · GCP · ETL · Data Warehouse

MLOps & Observability: MLflow · Prometheus · Grafana · ELK stack · Splunk · CI/CD · Docker

Languages: Python (10yr) · C/C++ · SQL · Bash

Infrastructure: Kubernetes (K3s) · Linux kernel · Vagrant · GitHub Actions

GitHub Stats

GitHub Streak

Looking For

I'm actively seeking roles in AI Platform Engineering, MLOps, and Site Reliability Engineering — where system design and infrastructure reliability are first-class concerns.

📄 CV · 💼 LinkedIn · 📫 lea.yeh.ml@gmail.com

Pinned Loading

  1. MiniShell is a minimalistic shell interpreter, supporting features like basic command execution, environment variable handling, command chaining, redirections, pipes, subshells, and here documents.

    C 46 2

  2. Lightweight HTTP/1.1 Web Server — A C++ web server built from scratch, compliant with RFC 7230. Inspired by Nginx architecture, it supports static file serving, modular configuration parsing, and b…

    C++ 3

  3. This project is just for learning purpose and to practice designing a neural network framework from scratch.

    Jupyter Notebook 1

  4. A lightweight Python toolkit for machine learning preprocessing and utilities — inspired by scikit-learn, built for clarity and educational purposes.

    Python 1

  5. A modern C++ library built through 42 School exercises — practicing SOLID principles, hexagonal architecture, and architectural decision-making via ADRs.

    C++