Amir Sarrafzadeh Arasi
Amir Sarrafzadeh Arasi
About Me
Hello! I'm Amir Sarrafzadeh Arasi, a GIS & Software Engineer with 5+ years of experience building geospatial platforms, Python geoprocessing tools, and ArcGIS Enterprise extensions (SOI/SOE) for government and research clients. I'm currently based in Bern, Switzerland and open to relocation across Switzerland and the EU.
My current research focuses on AI-assisted urban planning, agent-based traffic simulation (MATSim), and 15-minute-city modeling at ZHAW's SmartUrbanity EU project, where I work on the Thurgau Use Case for Demand-Responsive Transport (DRT) integration with public transport. Alongside this, I continue to work with WhereTech (Rome) on enterprise geospatial platforms, giving me a cross-border Switzerland/Italy professional life.
Originally from Tabriz, Iran, I bring a rich multilingual background (Persian and Azerbaijani native, with C1 English, C1 Turkish, and B2 Italian) and a deep enthusiasm for technology that turns complex urban challenges into clean, usable software. My academic foundation in Transport Systems Engineering at Sapienza University of Rome shaped the way I think about mobility, simulation, and data-driven planning.
On the technical side, I work day-to-day with ArcGIS Pro, ArcGIS Enterprise SDK, ArcPy, Python on Linux and Windows, and road-network data across schemas (OSM, MATSim network XML, shapefile conversion). I'm passionate about combining generative AI, 3D, and simulation to shape the next generation of urban tools.
Outside of work I enjoy chess, reading, time in nature, and keeping disciplined with the gym. I believe staying calm and focused is key to solving problems effectively, and I'm always looking for new challenges that let me learn, build, and collaborate with great people.
Thank you for taking the time to read about me. Let's create something meaningful together!
Work Experience
University Research Assistant
ZHAW Zurich University of Applied Sciences | Winterthur, Switzerland | Oct 2025 - Present
- Run MATSim agent-based DRT simulations for Canton Thurgau as part of the SmartUrbanity EU project (15-minute city theme).
- Develop Python validation scripts comparing Swiss Microcensus mobility data against simulation outputs.
- Orchestrate large-scale MATSim simulations on Linux infrastructure (Hetzner Ubuntu, ZHAW SLURM HPC cluster).
- Build Python/GeoPandas pipelines for geospatial analysis of MATSim outputs, including MATSim network XML to ArcGIS Pro shapefile conversion.
- Coordinate SmartUrbanity WP6 dissemination activities and contribute to public-facing project deliverables.
Senior Software Engineer
WhereTech S.r.l. | Rome, Italy (Remote) | May 2021 - Present
- Architected end-to-end RESTful APIs and microservices (Python/FastAPI, Java/Spring Boot) serving enterprise clients, including national government geospatial platforms used across all regions of Italy.
- Designed event-driven pipelines with Apache Kafka and RabbitMQ to ingest, process, and route real-time GPS and IoT sensor streams, enabling sub-second geotrigger evaluation and automated alerting.
- Engineered ArcGIS Enterprise extensions — Server Object Interceptors (SOIs) and Extensions (SOEs) — using the ArcGIS Enterprise SDK (Java, Apache Maven, Apache Tomcat), delivering custom spatial analysis capabilities for enterprise clients.
- Developed and maintained GeoEvent Services for live asset tracking, integrating field device feeds into enterprise ArcGIS dashboards consumed by operational teams.
- Optimized geospatial databases (PostgreSQL/PostGIS, MySQL, MongoDB), authoring complex spatial queries and automated data-migration scripts supporting datasets of millions of geographic records.
- Applied machine learning models (Scikit-learn, Python) to classify transport modes and detect anomalous patterns in mobility datasets, improving operational decision-making accuracy.
- Delivered full project lifecycle in agile sprints: API design, coding, Pytest/Locust performance testing, Docker containerization, and production deployment on AWS (EC2, RDS, S3).
Guest Researcher
Technical University of Denmark (DTU) | Kongens Lyngby, Denmark | Dec 2020 - May 2021
- Developed deep learning models (TensorFlow, Python) for transport mode classification and stop/motion detection on real-world GPS trajectory datasets (GeoLife, GPS-GIS-MMM).
- Built interactive geospatial visualizations using Leaflet.js and GeoPandas to communicate findings to research stakeholders.
Junior Software Engineer (Internship)
Movesion S.r.l. | Rome, Italy | Sep 2019 - Dec 2020
- Built transport mode detection algorithms and ML models (Scikit-learn) to calculate CO₂ footprints for the Mobility City mobile application.
- Collected, preprocessed, and analyzed geospatial and smartphone sensor trajectory data for user behavior modeling pipelines.
Education Background
Master's in Transport Systems Engineering (110/110)
Sapienza University of Rome | Rome, Italy | Sep 2018 - Jul 2021
- Specialized in transportation systems, data analytics, and software development.
- Conducted research on sustainable transportation and smart mobility solutions.
- Developed a thesis focusing on integrating machine learning models for transport mode detection.
- Graduated with distinction (110/110).
Bachelor's in Civil Engineering
Azad University of Tabriz | Tabriz, Iran | Sep 2006 - Feb 2012
- Focused on concrete and steel structures, transportation planning, and road engineering.
- Acquired foundational knowledge in engineering principles and problem-solving techniques.
- Completed a capstone project on urban road network optimization and management.
Skills
GIS & Geospatial
ArcGIS Pro · ArcGIS Enterprise · ArcGIS Online · ArcGIS API for Python · ArcPy · ArcGIS Enterprise SDK (SOI/SOE) · GeoEvent Server · CityEngine / CGA (Currently learning) · MATSim · PostGIS · GeoPandas · Leaflet.js · OSM
Languages & Frameworks
Python (FastAPI, Django, Flask, ArcPy) · Java (Spring Boot) · Node.js · SQL · Go (Currently learning)
AI & Machine Learning
Scikit-learn · TensorFlow · PyTorch · Pandas · NumPy · Deep Learning · Time-Series Analysis
Architecture & Messaging
RESTful APIs · Microservices · Apache Kafka · RabbitMQ · Distributed Systems (Currently learning)
Databases
PostgreSQL / PostGIS · MySQL · MongoDB · Redis · Elasticsearch · SQL Server · Firebase · Supabase
Frontend & Visualization
Bootstrap · JavaScript · HTML5 · CSS3 · Plotly · Streamlit · Leaflet.js
Cloud, DevOps & Tooling
AWS (EC2, S3, Lambda, RDS, CloudWatch) · Docker · CI/CD (GitHub Actions) · Apache Tomcat · Apache Maven · Git · Kibana · Jira · Confluence · Linux · Certified Scrum Master (CSM)
Languages
Persian & Azerbaijani (Native) · English C1 · Turkish C1 · Italian B2
Projects
Thurgau DRT Traffic Simulation & 15-Minute-City Analysis
Context: ZHAW SmartUrbanity EU Project | Oct 2025 – Current
- Ran large-scale MATSim traffic simulations with Demand-Responsive Transport (DRT) fleets for Canton Thurgau, modeling on-demand mobility alongside public transport in rural Swiss areas.
- Built a Python validation suite benchmarking simulation outputs against Swiss Microcensus data across trip distance, duration, mode, and departure time.
- Developed GeoPandas pipelines to convert MATSim network XML into ArcGIS Pro shapefiles, aligning road graphs between MATSim, OSM, and ArcGIS.
- Orchestrated simulation runs on Linux infrastructure (Hetzner, ZHAW SLURM cluster) with a reproducible Python/Java/Maven stack for 15-minute-city studies under the EU DUT Partnership.
Technologies: MATSim, Python, GeoPandas, ArcGIS Pro, Linux, SLURM, Java, Maven
RNDT National Geospatial Metadata Portal – FASTWEB
Duration: Jun 2024 – Current
- Engineered RESTful APIs for Italy's National Catalog for Spatial Metadata (RNDT), enabling government users to query and download geospatial metadata across all Italian regions via Elasticsearch-powered full-text and spatial search.
- Designed the relational database schema for metadata ingestion and built an automated harvesting pipeline using Geoportal Server 3.0.1.
Technologies: Java, Apache Tomcat, Elasticsearch, MySQL, Maven, Kibana, Bitbucket, Jira
View ProjectReal-Time Truck Supervision Platform
Duration: Jan 2024 – Jun 2024
- Built a real-time vehicle tracking system with geotrigger rules dispatching instant alerts on zone breaches or abnormal stops, processing 1,000+ location events per hour.
- Integrated a RabbitMQ event bus between field GPS devices and the ArcGIS GeoEvent Server backend, ensuring zero-message-loss delivery.
Technologies: Python, GeoEvent Services, RabbitMQ, Java, Maven, ArcGIS, Git, Jira
Enel Meteorological Data Processing System
Duration: Feb 2022 – Feb 2023
- Designed a real-time meteorological data pipeline consuming feeds from the Meteomatics API via Apache Kafka; routed extreme-weather alerts to Enel asset managers via automated email and dashboard notifications.
- Built a Node.js frontend for operations teams to monitor live meteorological indicators across Enel asset locations.
Technologies: FastAPI, Apache Kafka, Java, GeoEvent Services, Node.js, HTML5, CSS, Apache Maven, Git
AI-Powered Urban Mobility Prediction Engine
Type: Research & Development
- Designed and trained a multi-class deep learning model to predict citywide transport demand patterns 30 minutes ahead, leveraging historical GPS trajectory data and real-time weather feeds.
- Exposed predictions via a FastAPI REST service containerized with Docker and deployed on AWS EC2, integrated into a Node.js dashboard for urban planners.
Technologies: Python, TensorFlow, FastAPI, PostgreSQL, Docker, AWS, Node.js
SmartUrbanity – Full-Stack Urban Data Platform
Context: ZHAW / DUT EU Project
- Built and deployed a full-stack DUT project platform from scratch: Node.js backend, Tailwind CSS responsive frontend, and Supabase relational database; live at smarturbanity.com.
- Integrated Google Analytics 4 and implemented CI/CD via Vercel for automated deployments on every push.
Technologies: Node.js, Supabase, Vercel, Tailwind CSS, Google Analytics 4
View ProjectGeospatial Data Pipeline & Tile Server
Type: Personal / Engineering
- Built an automated ETL pipeline to ingest, reproject, and validate multi-source vector/raster geospatial datasets (Shapefile, GeoJSON, GeoTIFF) into a PostGIS database, reducing data preparation time by ~70%.
- Deployed a GeoServer instance via Docker to serve standardized WMS/WFS endpoints, integrated with a FastAPI middleware layer for access control and request routing to downstream GIS applications.
Technologies: Python, PostGIS, FastAPI, GeoServer, Docker
ResJet – Petrol Distribution & Monitoring Web Application
Duration: Jul 2024 – Current
- Design and model database structures, including tables and schemas.
- Migrate data from old DB (SQL Server) to new DB (PostgreSQL).
- Ensure database design supports application and business processes.
Technologies: MySQL Server, PostgreSQL, DBeaver, Python, Jira, Git
Puglia Sewer Water Management
Duration: Apr 2023 – Feb 2024
- Map and analyze the sewer network and surrounding environment.
- Implement and monitor water treatment processes to ensure quality standards.
- Design and develop microservices to regularly check water levels and manhole situations.
- Develop geoprocessing tools to analyze data and send reports to clients.
Technologies: ArcGIS Pro, ArcGIS Utility Network, Python (FastAPI), RabbitMQ, PostgreSQL, Git, Jira
Public Administration Geo-Spatial Filter
Duration: Oct 2021 – Dec 2021
- Implement a Server Object Interceptor (SOI) for ArcGIS Server.
- Filter map service requests based on user group membership.
- Handle user authorization and token authentication.
Technologies: ArcGIS Enterprise SDK, Java, Apache Maven, Apache HttpComponents, OAuth 2.0, Git