Amazon ECR Experts in Germany

Designed and developed an AI-powered automation platform using n8n to analyze social media niches, identify target audiences, and automate marketing strategy generation. The solution combined AI agents, workflow orchestration, and data analysis to automate research processes and generate data-driven insights.

• Designed and implemented complex automation workflows using n8n
• Developed AI-powered analysis agents for market and audience research
• Integrated multiple APIs and AI services into automated workflows
• Built automated market, competitor, and target audience analysis pipelines
• Leveraged Large Language Models (LLMs) for information summarization, classification, and prioritization
• Containerized and deployed the platform using Docker

Technologies: n8n, AI Agents, OpenAI APIs, Prompt Engineering, LLMs, Docker, Linux, REST APIs, Webhooks

• Architected and deployed end-to-end machine learning pipelines across classification and prediction datasets, ensuring robustness and reproducibility through MLOps best practices.
• Contributed directly to LLM model output accuracy improvement by designing and engineering specialised prompts grounded in end-to-end ML and SciML pipeline logic.
• Developed training data for large language models by formulating coding problems that models could not resolve and subsequently documenting the correct solutions.

• Developed end-to-end reproducible ML pipelines (PyTorch) with data versioning (DVC), experiment tracking (MLflow), automated testing (PyTest), and CI/CD across all training workflows.
• Scaled Vision Transformer and CNN training across NVIDIA A100 GPU clusters (CUDA, DDP, SLURM); applied hyperparameter optimization (W&B Sweeps) to reduce training overhead and identify optimal configurations.
• Developed a real-time 3D human motion generation system (ViT, VQ-VAE, SMPL-X/PIXIE) for personality-conditioned avatar synthesis; achieved state-of-the-art FID = 6.15 and P-FID = 10.31 on the UDIVA benchmark.
• Validated model expressiveness through structured user studies, achieving 86% accuracy in distinguishing extroverted vs. introverted avatar behaviors.
• Optimized inference pipelines by deploying PyTorch models via TensorRT and ONNX Runtime into native C++ code; benchmarked performance.

• It is like a startup within REWE, where we have to build a new forecasting system on Google Cloud Platform from the scratch. Although, officially my role is called MLOps, my actual tasks also include development of data processing pipelines (data engineering) and data scientists tasks such as feature engineering and model trainings.
• GCP: Terraform (tofu), Vertex AI (Kubeflow), Cloud Run, IAM, Google Cloud Storage, BigQuery, Artifact Registry
• Data engineering: Snowflake as the main data warehouse, Terraform, DBT for data model implementations
• CI/CD: GitLab. We have built a CI/CD pipeline that automates deployments of new releases up to production environment

• Creation of medical templates according to ISO 13485:2016 and GMP / MDR.
• Design Qualification (DQ).
• Validation plan (VP).
• Creation of PFMEA / FMEA.
• Installation Qualification (IQ).
• Operational Qualification (OQ).
• Performance Qualification (PQ).
• Final report (AB).

• As part of a greenfield project, a new production plant is being built in Slovakia to produce industrial media storage for OEMs and distributors once completed.

• Procurement and industrialization of production equipment for prefabrication.

• Design, procurement, installation and integration of complete production lines based on stamping, forming and welding technology with integrated, labor-reducing processes.

• Establishment of a production environment under strategic aspects according to the latest state of the art.

• Overall and continuous communication with the entire greenfield team, adjacent departments and plant and executive management.

• Development of concepts together with key suppliers and identification of ratio potentials along manufacturing process chains using value stream analyses.

• Collection, assessment and application of process data such as throughput times, manufacturing times, setup times, wait and idle times using time and multimoment studies as well as surveys.

• Forward-looking design of production systems and processes with the goal of implementing automated and competitive process solutions as part of comprehensive supply chain management.

• Planning of production workflows, definition of manufacturing methods, design of the overall layout and workstations.

• Determination and recording of process cost rates, overhead rates, cost evaluation of production processes, machine hourly rates, depreciation as well as expected share of claims and rework costs to optimize manufacturing costs.

• Continuous optimizations of product designs for their adaptation into automation-oriented production environments until the plant's SOP.

• Multi-project management for the engineering, design, procurement, installation and process integration of all relevant trades with respect to plant engineering, lifting equipment, production equipment and QA installations.

• Ongoing establishment of an overall concept in constructive, plant engineering, logistical, IT and QA aspects.

• Creation, maintenance and hardening of Kubernetes clusters employing Ansible and ArgoCD
• Keywords: Ansible, AWX, Kubernetes, NetApp, Prometheus, CI/CD ArgoCD, SSO, Fluent-bit, HAProxy, Calico, Keycloak, oauth2-proxy, SealedSecrets, kubeseal, Aqua kube-bench, CIS-Benchmarks, Aqua Trivy operator

From Crude to Crafted: Refining Participatory Design Data into Stakeholder-Ready Outcomes

• Architected a production Document AI platform using Retrieval Augmented Generation (RAG) over 1,500+ participatory design artefacts to answer historical project queries with grounded responses.
• Designed LLM evaluation combining RAGAS, custom evaluation metrics and human-in-the-loop (HITL) validation workflows to evaluate factual grounding, response quality, and prompt performance.
• Built a React, TypeScript, and D3.js frontend for interactive exploration of AI-generated insights.
• Implemented input layer LLM safety controls and Guardrails, including PII redaction and foul language filtering.

Software solution company that offers tools for data-driven campaign management for Amazon Ads.

• From scratch, developed an insights tool for the Amazon DSP with several data pipelines to support reports
• Developed new backend features for the Ad Optimiser tool
• Developed and maintained data pipelines for the Ads reports
• Wrote unit tests and reviewed code from other developers
• Supported frontend developers and performed frontend development tasks

• Migrated the Electric Vehicle Charging SaaS App of the EV Division from on-premises and Azure to AWS Cloud, resulting in a hybrid multi-cloud multi-tenant solution
• Developed a streaming data pipeline using AWS MSK for Apache Kafka and implemented an event-driven architecture to ingest and process near real-time data from OCPI-protocol IoT devices
• Implemented multi-tenant strategies including database schema isolation, bridge model for resource sharing, and tenant-based RBAC controls
• Provisioned Kubernetes clusters on AWS EKS with namespaces and RBAC for tenant isolation
• Led migration from on-premises and Azure to AWS using AWS DataSync, Snowball, and Database Migration Service
• Orchestrated collaboration across 5+ systems, vendors, service providers, and on-site teams
• Supported development and maintenance of IT strategy aligned with business requirements
• Managed €40 million infrastructure budget with AWS & Azure cost optimization, achieving 15% savings
• Led 50+ developers to implement advanced database procedures, increasing productivity by 20%
• Spearheaded multi-cloud, multi-tenant infrastructure migration for 30% faster processing times
• Negotiated vendor pricing to reduce payroll/benefits administration costs by 20%
• Developed a two-year infrastructure technology roadmap yielding 25% cost savings
• Tech stack: Kubernetes on AWS EKS, Docker, Kafka/AWS MSK, Terraform, AWS CDK, TypeScript, React, NextJS, Node.js, NestJS, Python, Aurora Serverless, RDS (MySQL, SQL Server), GitHub Actions, Azure DevOps, ArgoCD, AWS Lambda, API Gateway, AWS Security Hub, AWS Database Migration Service, AWS DataSync, AWS Organizations, AWS Control Tower, Odoo, Microsoft Navision, MS Dynamics

• Gathered tire images and their feature descriptions.
• Cleaned dataset of image metadata using pandas.
• Stored image feature embeddings in Chroma vector db.
• Used image augmentations to increase dataset size.
• Used sklearn to create shuffled datasets and imbalanced-learn to balance class sizes in dataset.
• Used PyTorch to train and test different neural networks.
• Validated model using custom accuracy metric based on similarity search in ChromaDB.
• Visualized accuracy predictions using matplotlib.
• Plugged trained model into DreamBooth to train stable diffusion model and generate new images of tires.
• Created custom Docker image in Amazon Elastic Container Registry for machine learning script.

• Designed and led the development of an automated short-video generation system.
• Built a scalable AWS backend using Step Functions, Lambda, S3, ECS Fargate, and DynamoDB.
• Developed video rendering with OpenCV and FFMPEG; ensured maintainable Python code.
• Supervised and mentored a Python developer and trained the client in AI workflows.
• Decreased end-to-end production time from hours to minutes.
• Created a modular, extensible architecture designed to support future AI models.

• Technical project management from inquiry to series release
• Monitoring project progress in terms of cost, schedule and quality
• Design of deep-drawing tools and conceptualization of plastic parts and tooling
• Development of manufacturing concepts and coordination with specialist departments
• Performing feasibility analyses and cost calculations, technical support for proposals