Hi, I'm

Anton Wingeier

I built a Kubernetes cluster on Raspberry Pis that hosts every project on this site. No cloud, no monthly bills. Just spent the past year building real-time traffic systems for the City of Tallahassee.

Scroll

I just graduated from Florida State University with a BS in Computer Science, focused on software engineering and data systems. This September I'm starting an MSc in Data Science at SDU (University of Southern Denmark) in Kolding.

I built a Kubernetes cluster on Raspberry Pis that hosts every project on this site. No cloud, no monthly bills. Just spent the past year building real-time traffic systems for the City of Tallahassee.

I'm looking for student worker or part-time data and software roles in Denmark starting fall 2026, ideally in or commutable from Kolding. If something here looks like a fit, I'd love to talk.

Skills

Languages

Python C++ JavaScript TypeScript SQL Java

Web & Data

Next.js React Flask Dash Pandas PostgreSQL Prisma Power BI Tableau

Infrastructure

Kubernetes (k3s) Docker Linux Tailscale Cloudflare Git
May 2026 - Present

Data Engineer (Contract)

Flip My Life Wellness Inc. — Remote

  • Designing and implementing ETL pipelines (Python/SQL) integrating Shopify, Klaviyo, Amazon Seller Central, and Stripe into a centralized analytics warehouse for a multi-channel DTC subscription business.
  • Building automated Power BI dashboards covering marketing attribution, customer LTV and cohort analysis, and subscription churn metrics — replacing manual reporting workflows with self-service analytics.
PythonSQLETLPower BIShopify APIKlaviyoStripeData Warehouse
Aug 2025 - May 2026

Software Engineering Intern

City of Tallahassee, Tallahassee Advanced Traffic Management System (TATMS)

  • Built a full-stack traffic operations dashboard (Python/Dash) integrating live data from Waze, Miovision, and BlueToad APIs with real-time Google Maps visualization across 385+ intersections, reducing incident acknowledgement time by 42%.
  • Designed a PostgreSQL database ingesting real-time data from 5 APIs, scaling to 20M+ rows with parallel batch processing and conflict-safe upserts to maintain sub-minute refresh rates.
  • Built automated test suite for data ingestion pipelines after an upstream API change caused a dashboard outage; tests have since caught 3 breaking changes before reaching production.
  • Developed Power BI dashboards with custom DAX measures and SQL queries to analyze overnight sensor and signal red-time data across 385+ intersections, identifying 15+ sensor failures and timing anomalies before they impacted live operations.
PythonDashPostgreSQLPower BITableauGoogle Maps APITesting
Oct 2024 - Aug 2025

Educator

E2 Young Engineers Tallahassee

  • Delivered hands-on STEM lessons to 30+ K–12 students twice weekly, introducing circuits, microcontrollers, and programming in C++ and Python.
  • Designed project-based curriculum covering robotics, sensors, and embedded systems to build engineering intuition in young learners.
C++PythonMicrocontrollers
Jan 2024 - May 2024

Data Analytics Intern

LARJA Consulting LLC, Boca Raton, FL (Remote)

  • Delivered automated reporting pipelines for 3 client accounts, reducing manual analysis time and saving an estimated 2+ hours/week.
  • Automated business-performance dashboards using Python and Pandas.
  • Restructured client databases for efficient data visualization and pivot analyses.
PythonPandasSQLExcel

Click any card to view the project.

Tallahassee Traffic Dashboard

The city's traffic operations center had 385+ intersections and no single place to see them all. I built the dashboard they use daily, pulling live feeds from Waze, Miovision, BlueToad, INRIX, and KITS into a real-time Google Maps view. Parallel ingestion keeps everything fresh, and an ML layer predicts sensor failures before they cause blind spots. Cut incident acknowledgement time by 42%.

City staff sign-in required to view the live ops dashboard.

PythonDashPostgreSQLGoogle MapsMLWaze API

Cypherify

Interactive cryptography learning tool. Covers Caesar through RSA with live frequency analysis and a cipher auto-detector that identifies the algorithm from ciphertext. Also has an OpenAI teaching assistant for when the math stops making sense.

FlaskDashPythonOpenAI

GoodSeason

Tracks what produce is actually in season where you live, then builds meal plans around it. Started because I kept buying strawberries in December without thinking about it. Shows the environmental cost of out-of-season farming and shipping too. Has its own domain at goodseason.me.

Next.jsTypeScriptOpenAIPostgreSQL

Couples Challenge

My girlfriend and I needed a way to keep each other on track with daily, weekly, and monthly goals. Full auth, automated challenge resets via cron, shared stats page.

Next.js 14TypeScriptPostgreSQLPrismaNextAuth

VoiceCast

Paste a chapter, a script, whatever. GPT-4o picks out every character, figures out their personality, and maps them to distinct ElevenLabs voices. Hit generate and you get a full multi-voice audio cast. Basically the audiobook producer nobody asked for.

Next.js 16React 19GPT-4oElevenLabsTypeScript

NetGuardian

My Eero gives me a device list and that's about it. So I built an IDS that actually watches the network. ARP scans every two minutes, catches spoofing in real time, port scans anything it doesn't recognize. The Flipper Zero with an ESP32 Marauder dev board handles WiFi scanning over serial UART, flagging rogue APs and evil twins. Web dashboard for alerts, SQLite backend, runs 24/7 on the Pi.

PythonScapyFlaskSQLiteNmapFlipper ZeroESP32 Marauder

Self-Hosted Kubernetes Cluster

The site you're reading right now is running on a geographically redundant k3s Kubernetes cluster I built across two physical locations ~600 miles apart, mirroring the same multi-site resilience principles behind AWS availability zones, built on Raspberry Pis.

If one site loses power or connectivity, the other keeps everything online automatically. Cloudflare Tunnel connectors run at both locations, so failover is instant with no manual intervention. Sites are connected over a Tailscale WireGuard mesh VPN with cross-site pod networking over VXLAN, and all traffic is served through Cloudflare with no open ports and no cloud hosting bills.

A NAS handles persistent deep storage across the cluster, keeping data intact through node failures, redeployments, or anything else the cluster throws at it. The setup includes a 3-node etcd control plane for high availability, PostgreSQL with persistent storage, and a build-push-rollout pipeline for zero-downtime deployments across all nodes.

Cluster Specs

Hardware

Raspberry Pi 5 (8GB) 3x Raspberry Pi 4 (4GB) 1TB NAS 2 Sites (~600 mi apart)

Stack

k3s Docker Tailscale Cloudflare Tunnel Traefik PostgreSQL Flannel VXLAN

I'm always interested in challenging work and new problems to solve. If something here caught your eye, I'd love to hear from you.

Say Hello