Machine Learning, Data Engineering, fine-tuning, Large Language Models (LLM), PyTorch, MLOps, Python, AWS, Google Cloud, Azure
Aktualisiert am 01.04.2026
Profil
Freiberufler / Selbstständiger
Remote-Arbeit
Verfügbar ab: 01.04.2026
Verfügbar zu: 100%
davon vor Ort: 100%
Python
PyTorch
Machine Learning
Deep Learning
Transformers
Generative AI
LLM Fine-Tuning
RAG
Model Evaluation
TensorFlow
Data Engineering
AWS
Google Cloud
Azure
Docker
Kubernetes
MLOps
Git
Apache Spark
FastAPI
Pydantic
SQL
Postgres
German
Polish
native
English

Einsatzorte

Einsatzorte

Berlin (+50km)
Deutschland
möglich

Projekte

Projekte

10/2025 - today: Fine-tuning of Small Language Models (SLMs) on extended datasets

Customer: on request
Role: Co-Founder

Tasks:
(On request) is a platform I co-founded, focusing on fine-tuning small language models (SLMs) and automating data augmentation and preparation workflows. The system simplifies the preparation, augmentation, and optimization of datasets for instruction-based model training, enabling individuals and organizations to build efficient, domain-specific models without requiring in-depth machine learning expertise. My role encompasses the platform's architecture and the development of training pipelines that support scalable and reproducible experimentation.

Skills:
Small Language Models (SLM and LLM), Fine-Tuning, Data Augmentation, Distillation, MLOps, Model Training Pipelines, Python, PyTorch, TensorFlow, Cloud Infrastructure, Agents
2024 - today: Research Project: Drone Real Time Scheduler

Tasks:
DRTS is an ongoing research project investigating distributed, space-time-aware coordination of heterogeneous UAVs within a self-organising multi-agent system. My current work focuses on the computer vision layer that enables perception-driven scheduling decisions in dynamic aerial environments. Using Roboflow for iterative dataset curation and active learning, I fine-tune YOLO-based object detection models with PyTorch and TorchVision to adapt to evolving mission scenarios (e.g. smart city, mapping, surveillance contexts). Training is conducted on scalable GPU infrastructure, while inference is optimised with OpenCV and deployed on resource-constrained edge hardware such as the Jetson Orin Nano companion computer. The perception outputs are integrated into the distributed peer architecture via MAVLink and MAVSDK, enabling each node to incorporate real-time environmental understanding into decentralised space-time scheduling algorithms. The research explores robustness, latency-performance trade-offs, adaptive ontologies, and perception-aware coordination strategies in ad-hoc UAV networks.
2017 - today: Teaching assignment

Role: Trainer and Consultant leading Data Science, Machine Learning, Angular, React, Node.js, React Native and Spring Boot Workshops

Tasks:
  • I have conducted highly successful workshops for esteemed clients
  • Delivered over 60 intensive workshops on Data Science, Machine Learning, Python, React, Angular, Vue, TypeScript, JavaScript, and NestJS, training more than 500 software professionals across Europe in both public and corporate settings (remote and on-site).
2014 - today: Development and architecture of data-driven systems

Role: Independent Software Architect, Machine Learning Engineer & Data Scientist

Tasks:
  • I work as an independent software architect/engineer and machine learning engineer for leading companies as well as for numerous European startups and public sector clients.
  • Delivering full-cycle development and architecture for data-driven systems that combine applied machine learning, modern web technologies, and large-scale distributed infrastructures. My work bridges AI research and software engineering, translating data insights into deployable, scalable, and compliant products.
  • Selected Areas of Contribution
    • Machine Learning & AI Integration - Developed and deployed ML models for product classification, anomaly detection, and computer vision in production environments.
    • Data Engineering & Analytics - Designed and optimized data ingestion, transformation, and analytics pipelines; implemented structured data layers for business and scientific use.
    • Software & System Architecture - Built modular architectures enabling high-performance data exchange, GPU-accelerated processing, and real-time decision workflows.
    • Edge & Cloud Solutions - Engineered distributed applications for healthcare, finance, and logistics, including UAV coordination and IoT data management.
    • Leadership & Collaboration - Guided cross-functional teams in designing, implementing, and delivering secure and compliant digital platforms.

Skills:
Python, TensorFlow, PyTorch, Keras, Pandas, PostgreSQL/ PostGIS, C++, TypeScript, React, Next.js, GPU Computing, Cloud/ CI Pipelines, Data Visualization, Distributed Systems, Java/ Kotlin, Spring Boot
2014: Development of a simulation and testing framework

Role: Research Associate
Customer: German Aerospace Center (DLR), Braunschweig

Tasks:
Contributed to the development of a simulation and test framework for the European Train Control System (ETCS), a Europe-wide railway safety and signalling standard. Focused on automating test scenarios through constraint-based simulation and ensuring system interoperability across high-speed rail networks.

Skills:
Research, Simulation Frameworks, Constraint Systems, Transportation Safety, Automation
04/2013 - 12/2014: Development of the Water Software ecosystem

Role: Software Engineer (employed)
Customer: iSAtech GmbH (Berlin)

Tasks:
Development of the iSAtech Water software ecosystem, including distributed backend services, client applications, and embedded device interfaces. Delivered core components such as key-management servers, data synchronization modules, and automated code-generation frameworks, integrating hardware communication, cryptography, and business intelligence workflows.

Skills:
Software Architecture, Distributed Systems, C++, Java EE, REST APIs, PostgreSQL
03/2012 - 10/2012: Development of the Water Software ecosystem

Role: Software Engineer (Werkstudent)
Customer: iSAtech GmbH, Berlin

Tasks:
Development of the iSAtech Water software ecosystem, including distributed backend services, client applications, and embedded device interfaces. Delivered core components such as key-management servers, data synchronization modules, and automated code-generation frameworks, integrating hardware communication, cryptography, and business intelligence workflows.

Skills:
Software Architecture, Distributed Systems, C++, Java EE, REST APIs, PostgreSQL
10/2011 - 03/2012: Development of the Water Software ecosystem

Role: Software Engineer (Freelancer)
Customer: iSAtech GmbH, Berlin

Tasks:
Development of the iSAtech Water software ecosystem, including distributed backend services, client applications and embedded device interfaces. Delivered core components such as key-management servers, data synchronization modules, and automated code-generation frameworks, integrating hardware communication, cryptography and business intelligence workflows.

Skills:
Software Architecture, Distributed Systems, C++, Java EE, REST APIs, PostgreSQL
03/2010 - 03/2012: various

Role: Software Engineer (Werkstudent)
Customer: IT SERVICE OMIKRON (ITSO) GmbH

Tasks:
IContributed to large-scale public sector systems including the hochschulstart.de admissions platform and the Visite suite developed for the Deutscher Bundestag. Transitioned from QA to software engineering, working on distributed, security-sensitive Java EE architectures integrating custom constraint solvers and REST-based communication.

Skills:
Java EE, Distributed Systems, Public Sector Software, PostgreSQL, Software Quality

Aus- und Weiterbildung

Aus- und Weiterbildung

7 Jahre 1 Monat
2005-10 - 2012-10

Study - Computer Science

Berlin Institute of Technology (TU Berlin)
Berlin Institute of Technology (TU Berlin)
Diploma Thesis: on request
1 Jahr 11 Monate
2002-08 - 2004-06

Abitur

Gymnasiale Oberstufe at OSZ Energietechnik in Berlin physics and electrical engineering
Gymnasiale Oberstufe at OSZ Energietechnik in Berlin physics and electrical engineering

Position

Position

Lead Machine Learning Engineer - Modern C++, Python, PyTorch, TensorFlow, Keras, LLMs (fi ne-tuning), Agentic Systems, React, Java, Kotlin

Kompetenzen

Kompetenzen

Top-Skills

Python PyTorch Machine Learning Deep Learning Transformers Generative AI LLM Fine-Tuning RAG Model Evaluation TensorFlow Data Engineering AWS Google Cloud Azure Docker Kubernetes MLOps Git Apache Spark FastAPI Pydantic SQL Postgres

Produkte / Standards / Erfahrungen / Methoden

Profile
  • I graduated with a degree in Computer Science from the Technical University of Berlin (TU Berlin). My focus was on distributed systems and software development. I worked as a software developer on projects for the visitor services of the German Bundestag and in the field of distributed systems. In addition to my technical skills, I have extensive knowledge of project and quality management. I gained my first research experience at the German Aerospace Center (DLR). I have also published several articles.
  • My work focuses on machine learning (C++, Python, PyTorch, and TensorFlow with Keras) and computer vision (OpenCV). I gained my first professional experience with TensorFlow and Keras during the development of the "Product Classification Tool (PCT)." This globally used tool assists customs authorities in classifying products based on an input vector and a machine learning model.
  • As a Data Engineer in the Promise Prediction Team, I supported ROSEN during the initial development of their computer vision-based classification tool as part of the Architectural Group and in building data classification services.
  • I co-founded the AI ??startup (platform for fine-tuning small language models - SLMs and data augmentation and distillation) - the platform is a "passion project".

Einsatzorte

Einsatzorte

Berlin (+50km)
Deutschland
möglich

Projekte

Projekte

10/2025 - today: Fine-tuning of Small Language Models (SLMs) on extended datasets

Customer: on request
Role: Co-Founder

Tasks:
(On request) is a platform I co-founded, focusing on fine-tuning small language models (SLMs) and automating data augmentation and preparation workflows. The system simplifies the preparation, augmentation, and optimization of datasets for instruction-based model training, enabling individuals and organizations to build efficient, domain-specific models without requiring in-depth machine learning expertise. My role encompasses the platform's architecture and the development of training pipelines that support scalable and reproducible experimentation.

Skills:
Small Language Models (SLM and LLM), Fine-Tuning, Data Augmentation, Distillation, MLOps, Model Training Pipelines, Python, PyTorch, TensorFlow, Cloud Infrastructure, Agents
2024 - today: Research Project: Drone Real Time Scheduler

Tasks:
DRTS is an ongoing research project investigating distributed, space-time-aware coordination of heterogeneous UAVs within a self-organising multi-agent system. My current work focuses on the computer vision layer that enables perception-driven scheduling decisions in dynamic aerial environments. Using Roboflow for iterative dataset curation and active learning, I fine-tune YOLO-based object detection models with PyTorch and TorchVision to adapt to evolving mission scenarios (e.g. smart city, mapping, surveillance contexts). Training is conducted on scalable GPU infrastructure, while inference is optimised with OpenCV and deployed on resource-constrained edge hardware such as the Jetson Orin Nano companion computer. The perception outputs are integrated into the distributed peer architecture via MAVLink and MAVSDK, enabling each node to incorporate real-time environmental understanding into decentralised space-time scheduling algorithms. The research explores robustness, latency-performance trade-offs, adaptive ontologies, and perception-aware coordination strategies in ad-hoc UAV networks.
2017 - today: Teaching assignment

Role: Trainer and Consultant leading Data Science, Machine Learning, Angular, React, Node.js, React Native and Spring Boot Workshops

Tasks:
  • I have conducted highly successful workshops for esteemed clients
  • Delivered over 60 intensive workshops on Data Science, Machine Learning, Python, React, Angular, Vue, TypeScript, JavaScript, and NestJS, training more than 500 software professionals across Europe in both public and corporate settings (remote and on-site).
2014 - today: Development and architecture of data-driven systems

Role: Independent Software Architect, Machine Learning Engineer & Data Scientist

Tasks:
  • I work as an independent software architect/engineer and machine learning engineer for leading companies as well as for numerous European startups and public sector clients.
  • Delivering full-cycle development and architecture for data-driven systems that combine applied machine learning, modern web technologies, and large-scale distributed infrastructures. My work bridges AI research and software engineering, translating data insights into deployable, scalable, and compliant products.
  • Selected Areas of Contribution
    • Machine Learning & AI Integration - Developed and deployed ML models for product classification, anomaly detection, and computer vision in production environments.
    • Data Engineering & Analytics - Designed and optimized data ingestion, transformation, and analytics pipelines; implemented structured data layers for business and scientific use.
    • Software & System Architecture - Built modular architectures enabling high-performance data exchange, GPU-accelerated processing, and real-time decision workflows.
    • Edge & Cloud Solutions - Engineered distributed applications for healthcare, finance, and logistics, including UAV coordination and IoT data management.
    • Leadership & Collaboration - Guided cross-functional teams in designing, implementing, and delivering secure and compliant digital platforms.

Skills:
Python, TensorFlow, PyTorch, Keras, Pandas, PostgreSQL/ PostGIS, C++, TypeScript, React, Next.js, GPU Computing, Cloud/ CI Pipelines, Data Visualization, Distributed Systems, Java/ Kotlin, Spring Boot
2014: Development of a simulation and testing framework

Role: Research Associate
Customer: German Aerospace Center (DLR), Braunschweig

Tasks:
Contributed to the development of a simulation and test framework for the European Train Control System (ETCS), a Europe-wide railway safety and signalling standard. Focused on automating test scenarios through constraint-based simulation and ensuring system interoperability across high-speed rail networks.

Skills:
Research, Simulation Frameworks, Constraint Systems, Transportation Safety, Automation
04/2013 - 12/2014: Development of the Water Software ecosystem

Role: Software Engineer (employed)
Customer: iSAtech GmbH (Berlin)

Tasks:
Development of the iSAtech Water software ecosystem, including distributed backend services, client applications, and embedded device interfaces. Delivered core components such as key-management servers, data synchronization modules, and automated code-generation frameworks, integrating hardware communication, cryptography, and business intelligence workflows.

Skills:
Software Architecture, Distributed Systems, C++, Java EE, REST APIs, PostgreSQL
03/2012 - 10/2012: Development of the Water Software ecosystem

Role: Software Engineer (Werkstudent)
Customer: iSAtech GmbH, Berlin

Tasks:
Development of the iSAtech Water software ecosystem, including distributed backend services, client applications, and embedded device interfaces. Delivered core components such as key-management servers, data synchronization modules, and automated code-generation frameworks, integrating hardware communication, cryptography, and business intelligence workflows.

Skills:
Software Architecture, Distributed Systems, C++, Java EE, REST APIs, PostgreSQL
10/2011 - 03/2012: Development of the Water Software ecosystem

Role: Software Engineer (Freelancer)
Customer: iSAtech GmbH, Berlin

Tasks:
Development of the iSAtech Water software ecosystem, including distributed backend services, client applications and embedded device interfaces. Delivered core components such as key-management servers, data synchronization modules, and automated code-generation frameworks, integrating hardware communication, cryptography and business intelligence workflows.

Skills:
Software Architecture, Distributed Systems, C++, Java EE, REST APIs, PostgreSQL
03/2010 - 03/2012: various

Role: Software Engineer (Werkstudent)
Customer: IT SERVICE OMIKRON (ITSO) GmbH

Tasks:
IContributed to large-scale public sector systems including the hochschulstart.de admissions platform and the Visite suite developed for the Deutscher Bundestag. Transitioned from QA to software engineering, working on distributed, security-sensitive Java EE architectures integrating custom constraint solvers and REST-based communication.

Skills:
Java EE, Distributed Systems, Public Sector Software, PostgreSQL, Software Quality

Aus- und Weiterbildung

Aus- und Weiterbildung

7 Jahre 1 Monat
2005-10 - 2012-10

Study - Computer Science

Berlin Institute of Technology (TU Berlin)
Berlin Institute of Technology (TU Berlin)
Diploma Thesis: on request
1 Jahr 11 Monate
2002-08 - 2004-06

Abitur

Gymnasiale Oberstufe at OSZ Energietechnik in Berlin physics and electrical engineering
Gymnasiale Oberstufe at OSZ Energietechnik in Berlin physics and electrical engineering

Position

Position

Lead Machine Learning Engineer - Modern C++, Python, PyTorch, TensorFlow, Keras, LLMs (fi ne-tuning), Agentic Systems, React, Java, Kotlin

Kompetenzen

Kompetenzen

Top-Skills

Python PyTorch Machine Learning Deep Learning Transformers Generative AI LLM Fine-Tuning RAG Model Evaluation TensorFlow Data Engineering AWS Google Cloud Azure Docker Kubernetes MLOps Git Apache Spark FastAPI Pydantic SQL Postgres

Produkte / Standards / Erfahrungen / Methoden

Profile
  • I graduated with a degree in Computer Science from the Technical University of Berlin (TU Berlin). My focus was on distributed systems and software development. I worked as a software developer on projects for the visitor services of the German Bundestag and in the field of distributed systems. In addition to my technical skills, I have extensive knowledge of project and quality management. I gained my first research experience at the German Aerospace Center (DLR). I have also published several articles.
  • My work focuses on machine learning (C++, Python, PyTorch, and TensorFlow with Keras) and computer vision (OpenCV). I gained my first professional experience with TensorFlow and Keras during the development of the "Product Classification Tool (PCT)." This globally used tool assists customs authorities in classifying products based on an input vector and a machine learning model.
  • As a Data Engineer in the Promise Prediction Team, I supported ROSEN during the initial development of their computer vision-based classification tool as part of the Architectural Group and in building data classification services.
  • I co-founded the AI ??startup (platform for fine-tuning small language models - SLMs and data augmentation and distillation) - the platform is a "passion project".

Vertrauen Sie auf Randstad

Im Bereich Freelancing
Im Bereich Arbeitnehmerüberlassung / Personalvermittlung

Fragen?

Rufen Sie uns an +49 89 500316-300 oder schreiben Sie uns:

Das Freelancer-Portal

Direktester geht's nicht! Ganz einfach Freelancer finden und direkt Kontakt aufnehmen.