Senior Machine Learning Engineer
Public Cloud Group GmbH
(500-1000 Mitarbeiter)
Schlagwörter
Architektur
Continuous Integration
Machine Learning
Chatbots
Large Language Models
APIs
Künstliche Intelligenz
Amazon S3
Data Analysis
Cloud Computing
Skills
Machine Learning und KI
Umfassende Erfahrung in der Entwicklung und Implementierung von Large Language Models (LLMs), Conversational AI und Retrieval-Augmented Generation (RAG).
Cloud-Technologien
Tiefgreifende Expertise in AWS-Services wie Lambda, S3, DynamoDB, ECR und Elastic Beanstalk. Erfahrung mit serverless Architekturen und Container-Technologien wie Docker.
Programmiersprachen und Frameworks
Fortgeschrittene Kenntnisse in Python, PySpark, SQL und verschiedenen ML-Frameworks. Erfahrung in der Entwicklung von API-Gateways und der Nutzung von Git für Versionskontrolle.
Data Science und Analyse
Fähigkeiten in statistischer Analyse, A/B-Testing, Zeitreihenvorhersage und explorativer Datenanalyse (EDA).
Software Engineering
Erfahrung in API-Design, serverlosen Architekturen, OOP, agilen Frameworks und skalierbaren Systemdesigns.
DevOps und CI/CD
Kenntnisse in der Implementierung von CI/CD-Pipelines mit GitLab und der Integration kontinuierlicher Bereitstellungsprinzipien.
Natural Language Processing
Erfahrung in der Entwicklung von NLP-Modellen, Textklassifikation und der Erstellung von AI-gestützten Übersetzungsplattformen.
Umfassende Erfahrung in der Entwicklung und Implementierung von Large Language Models (LLMs), Conversational AI und Retrieval-Augmented Generation (RAG).
Cloud-Technologien
Tiefgreifende Expertise in AWS-Services wie Lambda, S3, DynamoDB, ECR und Elastic Beanstalk. Erfahrung mit serverless Architekturen und Container-Technologien wie Docker.
Programmiersprachen und Frameworks
Fortgeschrittene Kenntnisse in Python, PySpark, SQL und verschiedenen ML-Frameworks. Erfahrung in der Entwicklung von API-Gateways und der Nutzung von Git für Versionskontrolle.
Data Science und Analyse
Fähigkeiten in statistischer Analyse, A/B-Testing, Zeitreihenvorhersage und explorativer Datenanalyse (EDA).
Software Engineering
Erfahrung in API-Design, serverlosen Architekturen, OOP, agilen Frameworks und skalierbaren Systemdesigns.
DevOps und CI/CD
Kenntnisse in der Implementierung von CI/CD-Pipelines mit GitLab und der Integration kontinuierlicher Bereitstellungsprinzipien.
Natural Language Processing
Erfahrung in der Entwicklung von NLP-Modellen, Textklassifikation und der Erstellung von AI-gestützten Übersetzungsplattformen.
Projekthistorie
03/2024
-
03/2025
• Led the development of a RAG knowledge platform using AWS CDK, Lambda,
Cognito, API Gateway, S3, and Bedrock, optimizing OpenSearch with LLM-based
semantic splitting, resulting in significant time savings for marketing and medical
professionals.
• Built a Python package for extracting predefined keywords from candidate resumes,
utilizing Bedrock, Jinja, and YAML, enhancing candidate selection efficiency and
quality.
• Directed a PoC email classification platform using Amazon Comprehend, finetuning
a custom classification model to achieve 92% accuracy, and implemented
automated endpoint management for cost savings.
Cognito, API Gateway, S3, and Bedrock, optimizing OpenSearch with LLM-based
semantic splitting, resulting in significant time savings for marketing and medical
professionals.
• Built a Python package for extracting predefined keywords from candidate resumes,
utilizing Bedrock, Jinja, and YAML, enhancing candidate selection efficiency and
quality.
• Directed a PoC email classification platform using Amazon Comprehend, finetuning
a custom classification model to achieve 92% accuracy, and implemented
automated endpoint management for cost savings.
07/2023
-
12/2023
Chief AI Officer
Cultway (Zapdeck) GmbH
• Formulated and executed a comprehensive AI strategy, developing and deploying an
OpenAI-based large language model architecture on Heroku to significantly
enhance the generation of high-quality presentation slides from text inputs.
• Fine-tuned LLM modules using synthetically generated data, resulting in faster and more
reliable outputs.
OpenAI-based large language model architecture on Heroku to significantly
enhance the generation of high-quality presentation slides from text inputs.
• Fine-tuned LLM modules using synthetically generated data, resulting in faster and more
reliable outputs.
07/2022
-
11/2023
Data & Software Engineer - Data & Analytics
AllCloud GmbH
• Led a team in developing a data platform utilizing AWS services (API Gateway, S3, Redshift,
Kinesis, Glue, SQS, SNS) and Apache Airflow. Ensured seamless orchestration via AWS
StepFunctions and AWS CDK, while integrating continuous integration and deployment (CI/CD) principles with GitLab.
• Developed an asynchronous processing system for real-time language translation,
leveraging AWS Transcribe, Translate, and Polly to manage continuous audio streams.
• Co-created an AI-translation platform, leveraging AWS services and SageMaker Augmented
AI, to create an AI-empowered translation solution. The solution automated translations,
reducing manual effort and costs, while maintaining quality through human review of
translations.
Kinesis, Glue, SQS, SNS) and Apache Airflow. Ensured seamless orchestration via AWS
StepFunctions and AWS CDK, while integrating continuous integration and deployment (CI/CD) principles with GitLab.
• Developed an asynchronous processing system for real-time language translation,
leveraging AWS Transcribe, Translate, and Polly to manage continuous audio streams.
• Co-created an AI-translation platform, leveraging AWS services and SageMaker Augmented
AI, to create an AI-empowered translation solution. The solution automated translations,
reducing manual effort and costs, while maintaining quality through human review of
translations.
10/2021
-
06/2022
Machine Learning Consultant - AI & Data Analytics
Deloitte GmbH
• Developed and automated a SageMaker-based recommender model pipeline, leveraging
CDK, Step Functions, S3, and ECR. Enabled model retraining and deployment on new data,
with Lambda for performance checks and automated endpoint updates for real-time
application recommendations.
• Executed a ransomware identification POC using software log data. Managed data
preparation, feature engineering, and EDA for time-series datasets. Developed and refined
LR and XGBoost models. Enhanced XGBoost accuracy to 90%, F1-Score to 0.87 and ROCAUC
to 0.93. Implemented cross-validation and hyperparameter tuning to mitigate
overfitting.
CDK, Step Functions, S3, and ECR. Enabled model retraining and deployment on new data,
with Lambda for performance checks and automated endpoint updates for real-time
application recommendations.
• Executed a ransomware identification POC using software log data. Managed data
preparation, feature engineering, and EDA for time-series datasets. Developed and refined
LR and XGBoost models. Enhanced XGBoost accuracy to 90%, F1-Score to 0.87 and ROCAUC
to 0.93. Implemented cross-validation and hyperparameter tuning to mitigate
overfitting.
01/2020
-
04/2021
Data Scientist
Axel Springer SE (idealo internet GmbH)
• Ensured the meeting of plan values for overall marketing and advertising campaigns
with time-series predictions embedded in Tableau with Python.
• Led A/B testing initiatives, designing test groups, monitoring systems, and providing
statistical evaluations using Bayesian models.
with time-series predictions embedded in Tableau with Python.
• Led A/B testing initiatives, designing test groups, monitoring systems, and providing
statistical evaluations using Bayesian models.
Reisebereitschaft
Weltweit verfügbar
