I'm Jakub Waller.
I'm a Hamburg-based Data Scientist, working in climate, at shipzero. Because there is no Planet B.
In my free time, you can catch me cycling Planet A, exploring its climate in various warming scenarios (+1.5°, +2.0°…) using my Probable Futures Bot, coding civic Telegram bots, reading an interesting book, or playing the guitar.
A short story so far.
Jakub Waller
Photo: studioline Photostudios, Leipzig
For the past 2+ years I've been a Data Scientist at shipzero, helping decarbonise global supply chains — building emission models, ML pipelines and data products that turn messy logistics data into actionable carbon insights. Because there is no Planet B.
Before that, I spent five years at Qimia GmbH on a wide range of Data Science, Machine Learning and Data Engineering projects, then took a year off to travel around our beautiful Planet A. That trip is what convinced me to dedicate the rest of my career to climate and data.
I graduated from the Czech Technical University in Prague with a master's thesis on time series classification using artificial neural networks, with exchange semesters at the University of Waterloo (Canada) and TalTech (Estonia). Earlier roles include Mibcon a.s. (SAP portal apps in Java/AngularJS) and bioinformatics work at the National Institute of Public Health.
University of Waterloo,
TalTech – Tallinn University of Technology
Career, education & craft.
Data Scientist
Decarbonising global supply chains. Building emission models, ML pipelines and data products that turn messy logistics data into actionable carbon insights for shippers and carriers.
Data Scientist
Working on various projects involving Data Science, Machine Learning and Data Engineering topics using a wide range of technologies.
Research Scientist
Comparing various architectures of artificial neural networks (convolutional neural networks, long short-term memory neural networks, and gated recurrent units) on a set of time series classification data sets in terms of the ability to learn, the effectivity of the training process, and the classification performance.
Java & JavaScript Developer
Programming SAP portal applications in Java, AngularJS and HTML for a web portal for ČEZ Distribuce, a.s.
Network Administrator
Maintenance of computers and other technical devices at the Institute of Immunology and Microbiology.
Bioinformatics Analyst
Extracting information from rRNA using various bioinformatics software.
Master of Science in Informatics
Main Topics: Pattern Recognition, Data Mining Algorithms, Data Preprocessing, Enterprise Data Warehouse Systems, Parallel Algorithms
Master’s Thesis: “Time Series Classification with Artificial Neural Networks” (B/1.5)
Exchange semester
Erasmus programme
Bachelor of Science in Informatics
Main Topics: Programming Languages and Compilers, Algorithms, Operating Systems, Database Systems, Security, Artificial Intelligence
Bachelor’s Thesis: “Simulation of a Quantum Particle on a Twisted 2D Waveguide” (B/1.5)
34 Things to Know about Carbon and Climate
How to Get a Job in Climate
ChatGPT Prompt Engineering for Developers
Data Visualization
Deploying Machine Learning Models in Production
Machine Learning Engineering for Production (MLOps)
Machine Learning Modeling Pipelines in Production
Machine Learning Data Lifecycle in Production
Introduction to Machine Learning in Production
Microsoft Certified: Azure AI Fundamentals
Side projects & tinkering.
A small collection of bots, tools and weekend builds — mostly Python, mostly on Telegram, mostly for fun.
Bürgerwecker
A free email service that watches Leipzig’s Bürgerbüro and Ausländerbehörde booking sites and notifies you the moment a matching citizen-office appointment or residence-document pickup slot opens up. Self-hosted on a Raspberry Pi with Flask, a Python poller and Docker Compose — no accounts, no tracking, no ads.
PapaMap
A map of ~6,000 places across Germany with a baby changing table, colored by whether a dad can actually reach it: green means an accessible room, red means women’s room only, grey means nobody has recorded which room yet. Grey pins deep-link to MapComplete, turning each gap into an OpenStreetMap edit. Rebuilt nightly from OSM on a Raspberry Pi.
Weather Analysis
Collects outside weather (Open-Meteo) and Home Assistant Zigbee sensor readings into a SQLite database on a Raspberry Pi, with a Streamlit dashboard for inside-vs-outside trends, temperature deltas, daily ranges and hour-by-day heatmaps — and how strongly (and how slowly) the flat follows the weather.
Zapfkompass
A map of 9,000+ German drinking venues, filterable by draft beer brand and serving type (Fassbier/Tankbier). Hamburg and Leipzig are swept in full via OpenStreetMap; venues elsewhere show up once a brand is known. Community submissions land on a human-curated core, refreshed nightly on a Raspberry Pi.
Probable Futures Bot
Probable Futures Bot is a Telegram Bot that is connected to the Probable Futures API and sends Probable Future scenarios for either an address or a location.
Covid Vaccination Appointments Bot
This bot was notifying users about free covid vaccination appointments in four cities: Hamburg, Leipzig, Stuttgart, and Dresden.
Strava Bot
Strava Bot sends you detailed statistics, graphs and maps after every ride/run/hike.
Resume
Who doesn't like recursion
Leipzig Appointments Bot
Tired of refreshing the page with available appointments at the Leipzig city office? The bot in this channel checks them for you and every minute sends available appointments.
Let's chat!
Curious about climate-tech, data, or just want to say hi? My inbox is open.