Fabian Crabus-GIS & AI Architect – Computer Vision and Geodata
Check rate
Experience
Senior GIS Developer
Transport & Logistics
Development of a route planner for incident communication.
- Development of the REST API
- Building a patch system for maintaining the routing graph
- Extending the test infrastructure
- Performance and memory optimization (JMeter, JFR)
Technologies: Java 21, Spring Boot, JGraphT, Flyway, MapStruct, Caffeine, ShedLock, JMeter, Kubernetes, JFR
Computer Vision Engineer
Internet & Information Technology
GPS-independent position determination of drones based on geocoded area data: building a pipeline from the segmentation of satellite and drone images through vectorization as 3D GeoJSON to polygon matching against a reference image.
- Object Segmentation of SAR data: detection of vegetation, buildings, and ditches
- Object Segmentation of EO drone data
- Conversion of segmentation results into 3D GeoJSON polygons
- Matching the current polygons against the reference overall image for localization
Technologies: U-Net, DINOv3, PyTorch
Short project: conversion of monocular images
- Conversion of monocular images into depth images and point clouds as training data for Jetson and Zed stereo cameras
- Development of a drone detection system based on audio and video using Python
Short project: AI for evaluating police reports
- Evaluation of police reports with LLMs and Named Entity Recognition
- Comparison of off-the-shelf open-source systems vs. fine-tuned models on an on-premise GPU cluster
Head of Development
Port Thessaloniki
To better predict container turnaround times in the port, measurement data (geodata, environmental data) is collected in real time via IoT sensors. The data flows, among other things, into an AI model based on time series so that logisticians can be given more accurate loading times.
- Technical leadership of the development team, architecture of the overall platform
- Real-time processing of geo and sensor data (Tile38, MQTT, Spark, Flink)
- AI model for time series prediction of container turnaround times (PyTorch, scikit-learn, DJL)
- Evaluation of edge/cloud approaches (AWS Greengrass, Azure IoT)
Technologies: Java 17/21, Quarkus (JPA, REST, Panache), Python, PyTorch, scikit-learn, DJL, Tile38, MQTT, PostgreSQL, GeoParquet, Spark, Flink, Superset, React, Ant Design, deck.gl, NodeJS, Docker, Kubernetes, OpenAPI, WebSockets, EDI, JUnit, Postman, nginx
Cloud: Azure Kubernetes Service, Azure AI, Azure ML
Environment: Scrum, Jira, Confluence, Bitbucket, Requirements Engineering
Short project: RAG-supported LLM
- In-house development of a Retrieval-Augmented-Generation LLM with Ollama and Langchain4j for domain-specific queries
Short project: Prototype analysis of ship sensor data
- Analysis of ship sensor data (position, draft, speed) for early wear detection of ship propellers with RNN models
- Predictive maintenance for maritime propulsion systems
Short project: AI models for audio data analysis
- Development of AI models for analyzing audio data with CNNs on FFT and STFT data
- Investigation of predictive maintenance for wear detection
Short project: AI training status quo
- Conducted a training session on the use of AI in Industry 4.0
Head of Development (AI / Big Data / Geodata)
Next Logistics Ventures GmbH (seabo.com)
"The Bloomberg Terminal for shipping": Building a startup for the digitalization of commercial shipping. The platform combines sensor data with a wide range of other data sources and gives ship owners and brokers deep insights into container and bulker shipping for better chartering deals.
Development of scalable microservices with Quarkus, Javalin, Spring Boot, and JBoss 7.4
Optimization of relational and NoSQL databases (PostgreSQL, MongoDB, DuckDB, Druid) with JPA/Hibernate/Panache
Integration of external APIs and ETL processes via REST, GraphQL, AWS Lambda, Spark, and Flink
Building a robust Kafka infrastructure and using AWS services (S3, Kinesis, SNS, SQS, SageMaker, Lambda)
Profiling, debugging, and optimization to improve software efficiency
Development and training of AI models for text processing (Bi-LSTM, LLM), anomaly detection (autoencoder), computer vision (YOLO, CNN, UNet), time series analysis (LSTM), and OCR tools (NER, LLM)
Processing and analysis of large data volumes (>50 GB/day) with Spark, Flink, and Druid
Development and optimization of ETL pipelines (NMEA AIS → Kafka → Flink → Parquet → Tile38 → AI model)
Development of geofencing toolchains, clustering analyses, data cleanup and downsampling methods, as well as models for distance calculation and route optimization
Visualization of geospatial data with QGIS, Kepler, ArcGIS, deck.gl, Python, and Spark
Technologies: Java 17 (Quarkus, Javalin, Spring Boot), Python, PyTorch, TensorFlow, Deeplearning4j, Langchain4j, Hugging Face Transformers, spaCy, OpenCV, CVAT, Tesseract, Kafka, Spark, Flink, Druid, Superset, PostgreSQL/PostGIS, MongoDB, DuckDB, Elasticsearch, GeoParquet, Avro, GraphQL, React, Ant Design, deck.gl, Plotly, D3, JTS, GeoTools, Spatial4j, Apache SIS, RasterIO, GeoPandas, Shapely, Pandas
Cloud: AWS (EC2, S3, Kafka MSK, Kinesis, SNS, SES, SQS), DigitalOcean, Imply Polaris, Jupyter
Environment: Requirements engineering, UML, C4, JUnit, JMeter, JProfiler, QGIS, ArcGIS, CI/CD, Jenkins, Docker, Kubernetes
Product Development
Meerkat.ai GmbH
- Development of a social media publishing platform for interacting with customer inquiries and complaints
- Assessing urgency using deep learning (RNN/LSTM)
- Development of a character-level GenAI chatbot
- Creation and implementation of the full software architecture
Head of Development
Viaboxx GmbH
- Building and leading a 12-person development team in the parcel automation environment
- Research projects on route optimization and load space monitoring
- International enterprise integration in Dubai, Denmark, and Norway (Camel, ServiceMix, SOAP, SAP JCO)
- Development of a work psychology assessment solution
- Creation and implementation of the software architecture
Head of Development
agimatec GmbH
- Building and leading a 6-person development team in the parcel automation environment with an international focus
- Use of Java 6, J2EE, Elasticsearch, JBoss 5, JMS, EJB3, JPA, Hibernate, ActiveMQ
- Implementation of ESB solutions (ServiceMix) and reporting (JasperReports)
- Use of Scrum for agile development
Technical Lead
DHL
- Development of a platform for controlling parcel lockers
- Creation and documentation of the architecture as well as implementation with Java 4, J2EE, EJB 2, Spring, JMS, and SonicMQ
- Use of rule engines (Drools), Oracle, and messaging solutions
- Conducting tests and documentation in the V-Model XT
Industry experience
See where this freelancer has spent most of their professional time.
Experienced in Transportation, Information Technology, Aerospace and Defense, and Government and Administration.
Business area experience
See which departments and functions this freelancer has contributed to most.
Experienced in Information Technology, Product Development, Research and Development, Logistics, Operations, and Business Intelligence.
Summary
Architect for geodata systems and computer vision with over 15 years of experience in logistics, IoT, and geoinformatics. My focus: AI models that work on spatial data – from segmentation of SAR and EO data (U-Net, DINOv3) to real-time geodata processing (Tile38, PostGIS, GeoParquet) and production-ready cloud architecture (AWS, Azure).
Skills
- Computer Vision & Ai: Object Segmentation (U-Net, Dinov3), Sar/Eo Data Analysis, Opencv, Cvat, Tesseract, Pytorch, Tensorflow, Deeplearning4j, Djl, Hugging Face Transformers, Llms (Langchain4j), Nlp (Spacy), Predictive Analytics, Time Series Analysis
- Geodata & Gis: Postgis, Tile38, Geoparquet, Geojson, Jts, Geotools, Spatial4j, Apache Sis, Rasterio, Geopandas, Shapely, Deck.Gl, Qgis, Arcgis, Routing Graphs (Jgrapht)
- Programming Languages: Java 17/21 (Quarkus, Spring Boot, Javalin), Python (Pandas, Scikit-Learn), Javascript (React, Nodejs)
- Software Architecture: Microservices, Rest, Graphql, Api Design, C4, Uml, Requirements Engineering
- Big Data & Streaming: Apache Spark, Flink, Kafka, Druid, Duckdb, Elasticsearch, Etl, Data Lakes, Superset
- Cloud & Devops: Aws (Ec2, S3, Kafka Msk, Kinesis, Sqs/Sns), Azure (Aks, Azure Ai, Azure Ml), Docker, Kubernetes, Ci/Cd (Jenkins, Gitlab), Maven
- Iot: Mqtt, Sensor Integration, Edge Computing (Aws Greengrass, Azure Iot), Edi
- Methods & Leadership: Scrum (Certified Scrum Master), Technical Leadership Of Teams Of Up To 12 People, Jira, Confluence
Languages
Certifications & licenses
BEA Weblogic Developer
Scrum Alliance
Statistics
Experience
Global experience
Expertise
Qualifications
Profile
Frequently asked questions
Have questions? Find more information here.
Fabian is based in Kalkar, Germany and prefers 100% remote projects.
Fabian speaks the following languages: German (Native), English (Advanced).
Fabian has at least 23 years of experience. During this time, Fabian has worked in at least 12 different roles and for 8 different companies. The average length of individual experience is 2 years and 8 months. Note that Fabian may not have shared all experience and actually has more experience.
Based on recent experience, Fabian would be well-suited for roles such as: Senior GIS Developer, Computer Vision Engineer, Short project: conversion of monocular images.
Fabian's most recent position is Senior GIS Developer at Transport & Logistics.
In recent years, Fabian has worked for Transport & Logistics, Internet & Information Technology, Port Thessaloniki, and Next Logistics Ventures GmbH (seabo.com).
Fabian is most experienced in industries like Transportation, Information Technology, and Aerospace and Defense. Fabian also has some experience in Government and Administration and Manufacturing.
Fabian is most experienced in business areas like Information Technology, Product Development, and Research and Development. Fabian also has some experience in Logistics, Operations, and Business Intelligence.
Fabian has recently worked in industries like Transportation, Information Technology, and Aerospace and Defense.
Fabian has recently worked in business areas like Information Technology, Product Development, and Research and Development.
Fabian has 2 certificates. These include: BEA Weblogic Developer and Certified ScrumMaster.
The availability of Fabian needs to be confirmed.
Daily rate distribution
The rates shown represent the typical market range for freelancers in this position based on recent contracts on our platform.
Average rates for similar positions
Rates are based on recent contracts and do not include FRATCH margin.
The average daily rate is the mean of all daily rates from recent contracts of comparable freelancers on our platform.
The median daily rate is the middle value of all daily rates — half of comparable freelancers charge less, half charge more. Unlike the average, it is barely affected by outliers.
Calculated based on our freelancers’ daily rates as of 7 Sep 2026. Actual rates may vary depending on seniority level, experience, skill specialization, project complexity, and engagement length.
Similar freelancers
Discover other experts with similar qualifications and experience
Jürgen Fey
Michael Fecher
Tobias Jaeuthe
Hans Gedon
Andreas Kraus
Ralph Elsäßer
Max Ritter
Michael König
Jörg Trappschuh
Louis Guitton
Experts recently working on similar projects
Freelancers with hands-on experience in comparable project as a Senior GIS Developer
Hooman Behmanesh
Ramazan Cinardere
Rainer Langbehn
Christoph Thodte
Jens Oberender
Ola Van Dunen
Florian Knopf
Jörg-Ulrich Hammerbacher
Florian Fladung
Thorsten Matzner
Nearby freelancers
Professionals working in or nearby Kalkar, Germany
Collin Kempkes
Gudrun Aydt
Fadi Shoaa
Felix Bauchspiess
Timo Bakenecker
Alwin Freelancer
Henry Hanau
Peter Sachs-Witzel
Rudolf Eggelbusch
Matthias Bullmahn
