About

Work Experience

• Senior Software Engineer, c-sense GmbH

  Aug, 2024 - Present

  c-sense develops nanoscale sensor technology and AFM/SPM instruments for scientific and industrial applications.

  • Driving the company's AI integration initiative as sole technical decision-maker — evaluating LLM tooling and defining the end-to-end agentic architecture strategy

  • Currently architecting an agentic RAG system (LangChain/LangGraph, Chroma, LLM cloud APIs) to give the CEO cross-domain visibility into internal technical documentation, sales/CRM data, competitor product analysis, and prospective customer profiling

  • Architected a modular, layered software framework for SPM instrument control — deliberately separating hardware vendor, orchestration, and application layers for long-term extensibility and reuse across future instrument variants

  • Defined cross-layer interface contracts and async command routing patterns to enable parallel development by hardware and software teams, reducing integration risk during concurrent R&D cycles

  • Led the design of a hardware abstraction strategy over a LabVIEW-based proprietary API, keeping the software stack agnostic to vendor-specific instrument changes

  • Delivered automated measurement capabilities (spatial sampling, real-time safety management) that allow researchers to focus on experimental outcomes rather than instrument operation

  • Collaborated with hardware engineers and research scientists to translate physics measurement protocols into reliable, automated software workflows

  • Built the GitOps delivery platform for the company's internal service pipeline — ArgoCD-driven continuous delivery across environments, with every architectural decision recorded as an ADR

• Data Scientist, MediaTek

  Sep, 2022 - Sep, 20231 year

  Devoted to data-related initiatives spanning cost control and chip development in the AI & Big Data department of a leading global semiconductor company.

  • Implemented model monitoring solutions using MLflow and Streamlit, ensuring high availability and performance consistency for production ML models

  • Implemented ML models to enhance mobile temperature control mechanisms, achieving temperature errors below 1°C

  • Collaborated with chip developers to reduce power consumption by 20%, resulting in a 14K Antutu score improvement in mobile performance

• Data Engineer, MediaTek

  Jul, 2019 - Sep, 20223 years 2 months

  Built data infrastructure and analytics platforms for one of the world's largest IC design companies.

  • Established automated data pipelines for structured and unstructured data; designed PB-level ETL processes using Airflow, NiFi, Dataflow, and BigQuery on GCP

  • Introduced dimensional-modeling and data-warehouse architecture, improving data table reuse rates during the company's early digital transformation

  • Implemented real-time data quality monitoring and alerting, reducing data missing rates from 50% to <1% and lowering monthly labor costs by 7.5 man-days

  • Built model monitoring infrastructure using MLflow and Grafana dashboards to track ML model performance and data drift in production

  • Managed EDA License and Computing Farm costs via interactive BI Dashboard (Splunk + Grafana), aiding EO procurement decisions and reducing costs by 25%

• Software Engineer, MediaTek

  Sep, 2016 - Jun, 20192 years 9 months

  • Developed debugging and analysis tools for Modem Logs

  • Developed an automated ICD DMS using Python and Jenkins to enforce documentation standards

  • Created a World Wide Field Trial Upload Tool using Vue.js with Electron

  • Developed a StackOverflow-like QA platform using AngularJS

Projects Experience

• lyeh-infra — Self-Hosted Kubernetes Infrastructure (GitOps)

  Apr, 2026 - Present

  Personal cloud infrastructure on a single Hetzner CX23 VM running k3s, deployed via ArgoCD GitOps. Everything is declared in Git — adding a new service means creating a directory and pushing; ArgoCD handles the rest. This portfolio site is hosted on it.

  • Designed a GitOps deployment model using an ArgoCD ApplicationSet with a Git directory generator — every subdirectory under apps/ is auto-discovered and deployed to its own namespace, with prune and self-heal enabled

  • Configured Traefik ingress (hostname-based routing) and cert-manager for automatic Let's Encrypt TLS certificate provisioning and renewal

  • Built a fully automated CI/CD pipeline (GitHub Actions): Hugo build → Docker image to GHCR pinned by content SHA → image tag written back into deployment.yaml → ArgoCD rolling update

  • Pinned deployments by immutable image SHA rather than mutable tags so Git remains an exact source of truth and ArgoCD reliably detects changes

  • Scripted the full zero-to-running bootstrap (provision → k3s → cert-manager → ArgoCD) so a fresh VM becomes self-managing in four steps

• ft_transcendence — Full-Stack Platform with Observability Stack

  Oct, 2025 - Mar, 20265 months

  Contributed the DevOps module for a full-stack multiplayer web platform (42 Vienna capstone project). Responsible for designing and deploying the complete observability infrastructure covering metrics, logging, and alerting.

  • Deployed Prometheus for metrics collection and configured alerting rules for service health and performance thresholds

  • Built Grafana dashboards for real-time visibility into application and infrastructure metrics across the platform

  • Set up ELK stack (Elasticsearch, Logstash, Kibana) for centralized log aggregation and search across distributed services

  • Containerized the full observability stack with Docker Compose, enabling reproducible deployment and local development parity

• ft_linux — Linux From Scratch

  Sep, 2025 - Jan, 20264 months

  Built a fully bootable Linux system from scratch, covering every layer from cross-compilation toolchain to kernel configuration, filesystem hierarchy, init system, and bootloader. Follows LFS/BLFS/ALFS methodology with a custom automation layer.

  • Compiled a custom Linux kernel (4.x) with hand-selected driver and filesystem configuration

  • Built a two-phase cross-compilation toolchain (temporary + final) to produce a host-independent, self-contained Linux system

  • Automated the full build pipeline (14 stages) via an ALFS-style bootstrap script with environment isolation and error recovery

  • Designed partition layout, configured GRUB bootloader, SysV init, and udev for dynamic device management

• Inception-of-Things — K3s Kubernetes Cluster

  Jun, 2025 - Oct, 20254 months

  Provisioned and configured a multi-node K3s (lightweight Kubernetes) cluster using Vagrant and VirtualBox, covering cluster networking, ingress, and GitOps-style deployment workflows.

  • Deployed a multi-node K3s cluster with Vagrant-provisioned VMs, including nested virtualization setup

  • Configured Kubernetes ingress, service routing, and workload deployment across cluster nodes

  • Applied GitOps principles with ArgoCD for continuous deployment in the bonus track

• libftpp — C++ Architecture Library

  Oct, 2025 - Jan, 20263 months

  A modern C++20 library built through structured exercises practicing SOLID principles, hexagonal architecture, and design patterns. Includes Architectural Decision Records (ADRs) documenting trade-off reasoning.

  • Implemented core data structures and utilities applying SOLID principles and hexagonal (ports & adapters) architecture

  • Documented architectural decisions via ADRs, treating design trade-offs as first-class engineering artifacts

  • Maintained CI pipeline with CMake, clang-format, and automated test suite

• litetorch + litetune + sklite — AI Framework Toolkit

  Mar, 2025 - Jun, 20253 months

  A trio of ML infrastructure projects — litetorch (neural network framework built from scratch, inspired by PyTorch), litetune (hyperparameter tuner inspired by Ray Tune), and sklite (ML preprocessing toolkit inspired by scikit-learn) — demonstrating ML internals depth beyond API usage.

  • Implemented forward/backpropagation, autograd, and layer abstractions from scratch in litetorch

  • Built a hyperparameter search and experiment tracking system in litetune, mirroring Ray Tune's trial management model

  • Designed preprocessing pipelines and utility functions in sklite for educational clarity and extensibility

• webserver — HTTP/1.1 Server in C++

  Jun, 2024 - Sep, 20243 months

  Implemented a standards-compliant HTTP/1.1 web server in C++ from scratch, handling concurrent connections, request parsing, and static/dynamic content serving.

  • Built non-blocking I/O event loop handling concurrent HTTP connections using poll/select

  • Implemented HTTP/1.1 request parsing, routing, and response generation

  • Supported CGI execution, static file serving, and configurable virtual hosts

• minishell — Bash-compatible Shell

  Dec, 2023 - Mar, 20243 months

  Reimplemented core GNU Bash shell features including command parsing, process management, and built-in execution.

  • Developed a syntax analyzer using the shift-reduce algorithm for Bash-like grammar interpretation

  • Optimized subprocess management and pipeline execution across multi-stage pipelines

  • Employed Docker to ensure consistent development environments across the team

• CDNJS — Content Delivery Network for JavaScript

  Sep, 2015 - Sep, 20161 year

  Open-source CDN for JavaScript libraries used by over 3.5 million websites, serving 30+ billion requests per month.

  • Contributed to a platform serving over 30 billion requests per month across 3.5 million websites

  • Assisted with library maintenance and automation tooling for the open-source CDN

Skills

Education

Certificates

Awards

Volunteer Work

Publications